阅读说明：本技术 一种用于商店内消费者行为事件元数据聚合、数据验证及其用于数据解释的人工智能分析和相关动作触发的系统
(System for in-store consumer behavior event metadata aggregation, data verification and artificial intelligence analysis and related action triggering for data interpretation
) 是由 迪佩什·艾夫拉尼
肖恩·塔赫尼
卡尔·吉尼 于 2018-06-22 设计创作，主要内容包括：本发明提供一种用于商店内消费者行为事件元数据聚合、数据验证及其用于数据解释的人工智能分析和相关动作触发的系统。所述系统可以从多个消费电子设备收集商店内消费者行为事件元数据,然后使用经过训练的人工智能分析引擎来提供对这些消费者有用的各种人工智能洞察,所述人工智能洞察可进一步修改消费者行为。所述经过训练的人工智能分析引擎可以具有数据解释控制器,所述数据解释控制器被配置为智能地解释此类聚合的商店内消费者行为事件元数据,并相应地触发动作,所述动作随后以电子方式发送到所述消费者电子设备。所述数据解释控制器可以具有数据验证控制器,所述数据验证控制器被配置成根据可用性和其它元数据优化表示多个消费品的消费品元数据数据库的数据完整性。(The invention provides a system for in-store consumer behavior event metadata aggregation, data verification and artificial intelligence analysis for data interpretation and related action triggering thereof. The system can collect in-store consumer behavior event metadata from multiple consumer electronic devices and then use trained artificial intelligence analysis engines to provide various artificial intelligence insights useful to those consumers that can further modify consumer behavior. The trained artificial intelligence analysis engine can have a data interpretation controller configured to intelligently interpret such aggregated in-store consumer behavior event metadata and trigger an action accordingly, which is then electronically transmitted to the consumer electronic device. The data interpretation controller may have a data validation controller configured to optimize data integrity of a consumer product metadata database representing a plurality of consumer products as a function of availability and other metadata.)
1. A system for in-store consumer behavior event metadata aggregation, data verification, and artificial intelligence analysis thereof for data interpretation and related action triggering, the system comprising:
a data model having:
consumer product metadata comprising at least:
consumer product ID metadata; and
consumer product location metadata;
the system is configured to:
receiving in-store consumer behavior event metadata from a plurality of consumer electronic devices, the in-store consumer behavior event metadata at least partially from a shopping list GUI displayed by a display device of each consumer electronic device, the shopping list GUI comprising consumer item lists specific to a consumer and associated purchase confirmation inputs indicating a purchase thereof,
inputting the in-store consumer behavioral event metadata into a trained artificial intelligence analysis engine, the artificial intelligence analysis engine trained using a machine learning algorithm having inputs as in-store consumer behavioral event metadata training data, and wherein the machine learning algorithm is trained at least from purchase confirmation data from the purchase confirmation inputs, the analysis engine comprising:
a data interpretation controller comprising:
data verification data interpretation controller, and
an action trigger controller comprising:
the data verification action controller is used for verifying the action of the data,
wherein the data verification data interpretation controller is configured to identify a data verification opportunity match from consumer behavioral event metadata within the store; and
generating a data verification action using the data verification action controller, the data verification action configured according to consumer behavioral event metadata within the store;
sending a data validation action electronic communication to at least one of the plurality of consumer electronic devices;
receiving product data metadata response data from the at least one shopper electronic device in response to the electronic communication of the data verification action; and
updating the consumer product data using the product data metadata response data.
2. The system of claim 1, wherein the data interpretation controller further comprises:
a product recommendation data interpretation controller, and wherein the action trigger controller further comprises:
a product recommendation action controller, and wherein, in use:
the product suggestion data interpretation controller is configured to identify a product suggestion opportunity match from the in-store consumer behavior event metadata; and
the product suggestion action controller is configured to generate and transmit a product suggestion action for display by the shopping list GUI.
3. The system of claim 2, wherein the product suggestion action includes in-store location data of the product suggestion.
4. The system of claim 1, wherein the in-store location data indicates at least one of an aisle number and a shelf.
5. The system of claim 3, wherein the electronic device further comprises a sensor for sensing in-store location data, wherein the in-store consumer behavior event metadata further comprises the in-store location data.
6. The system according to claim 5, wherein the sensor comprises at least one of a near field communication sensor and a Bluetooth Beacon (BLE) sensor.
7. The system of claim 1, wherein the electronic device further comprises an image sensor for capturing an image of the product, wherein the product data is from the image.
8. The system of claim 7, wherein the product data includes at least one of a product ID and product pricing data.
9. The system of claim 1, wherein the data interpretation controller further comprises:
the information informs the data interpretation controller; and wherein the action trigger controller further comprises:
the information informs the operation controller; and wherein, in use:
the information notification data interpretation controller is configured to identify and information notification opportunity matches based on the in-store consumer behavior event metadata, and
the information notification action controller is configured to send a notification to the at least one electronic device.
10. The system of claim 1, wherein the data interpretation controller is optimized using the machine learning algorithm.
11. The system of claim 1, wherein the action-triggering controller is optimized using the machine learning algorithm.
12. The system of claim 1, wherein the machine learning algorithm is configured to optimize the product suggestion controller according to purchase confirmation input data derived from the shopping list GUI.
13. The system of claim 1, wherein the machine learning algorithm is configured to optimize the data validation data interpretation controller according to a probability of receiving the product data metadata response data.
14. The system of claim 1, wherein the machine learning algorithm is configured to optimize the data validation data interpretation controller according to data integrity of the product data metadata response data.
15. The system of claim 1, wherein the machine learning algorithm is configured to optimize the information controller based on consumer interaction with information notification cues.
16. The system of claim 1, wherein the consumable metadata comprises at least one product category.
17. The system of claim 16, wherein the product data metadata response data includes product price metadata.
18. The system of claim 7, wherein the in-store consumer behavior event metadata comprises image data, and wherein the machine learning algorithm is configured to optimize the product suggestion action controller for object recognition of the image data.
Technical Field
The present invention relates generally to intelligent analytics engines, and more particularly to a system for in-store consumer behavior event metadata aggregation, data verification, and artificial intelligence analysis and related action triggering thereof for data interpretation, suitable for in-store applications in the real world.
Background
Consumer behavior data aggregation analysis and event triggering are common on the web. For example, various providers utilize tracking technologies (including the use of cookies) to track consumer behavior including consumer interests, which, after being determined, provide targeted advertising to the consumer accordingly.
However, such systems are not readily adapted for practical applications, such as for in-store consumer behavior analysis and consumer behavior modification.
Problems involved in seeking to deploy such analysis engines in the real world include consumer product metadata integrity issues, as products from different stores vary widely in availability and pricing.
Furthermore, there are difficulties in obtaining consumer behavior metadata from the real world.
Furthermore, prior art online targeted advertisement analysis engines typically only provide targeted advertisements that are selected from an advertisement database based on determined user interests. However, this would be an ideal choice for a system that is capable of doing more things, including generating further artificial intelligence insights that are useful to consumers in the real world.
The present invention seeks to provide a system and associated method which overcomes or substantially ameliorates at least some of the disadvantages of the prior art, or at least provides an alternative.
It is to be understood that, if any prior art information is referred to herein, that reference does not constitute an admission that the information forms part of the common general knowledge in the art, in australia or in any other country.
Disclosure of Invention
In the following embodiments, a system for in-store consumer behavior event metadata aggregation, data verification, and artificial intelligence analysis thereof for data interpretation and related action triggering is provided that overcomes or at least ameliorates the problems of prior art systems, or at least provides another option.
As will be described in further detail below, the system collects in-store consumer behavior event metadata from a plurality of consumer electronic devices and then uses a trained artificial intelligence analysis engine to provide these consumers with a variety of useful artificial intelligence insights to further modify consumer behavior.
As will be described in further detail below, the trained artificial intelligence analysis engine has a data interpretation controller configured to intelligently interpret such aggregated in-store consumer behavior event metadata and trigger actions accordingly, which are then electronically transmitted to the consumer electronic device.
In an embodiment, the data interpretation controller comprises a data validation controller configured to optimize data integrity of a consumer product metadata database representing a plurality of consumer products based on availability and other metadata (e.g., in-store location, pricing, etc.). Thus, the data verification controller is configured to first construct a relatively accurate metadata model of the consumer product in a manner to address the challenges of such data aggregation in the real world where the system may also make further artificial intelligence insights once such a relatively accurate data model is constructed.
In this regard, the system may further comprise an intelligent product recommendation function, wherein the data interpretation controller comprises a product recommendation controller.
The present system uses a shopping list Graphical User Interface (GUI) displayed via display devices of a plurality of consumer electronic devices, a shopping list GUI, a display configured to display a plurality of consumer-specific consumer items and a list format, wherein each consumer item displayed includes an associated purchase confirmation input (check box, etc.) indicating its purchase.
Thus, by using these consumer-specific shopping lists, particularly the consumer products listed above, the product suggestion controller is able to intelligently suggest products for consideration by the consumer.
In an embodiment, the product suggestion controller is capable of suggesting specific products within a product category in an intelligent manner, such as may be determined from consumer-specific parameters or obtained from machine learning of in-store consumer behavioral event metadata obtained by other consumers.
Further, in embodiments, the product suggestion controller can suggest products that appear to be unrelated but that may be related to the user, such as products identified through analysis performed by a machine learning algorithm.
Further, in embodiments, the trained artificial intelligence analysis engine is capable of generating intelligent consumer product information notifications that may be helpful to consumers. Such consumer product information notifications may further be derived from machine learning algorithms.
According to one aspect, there is provided a system for in-store consumer behavior event metadata aggregation, data verification, and artificial intelligence analysis thereof for data interpretation and related action triggering, the system comprising: a data model having: consumer product metadata, the consumer product metadata including at least: consumer product ID metadata; and consumer product location metadata, the system configured to: receiving in-store consumer behavioral event metadata from a plurality of consumer electronic devices, the in-store consumer behavioral event metadata at least partially from a shopping list GUI displayed by a display device of each consumer electronic device, the shopping list GUI comprising a list of consumer-specific consumer products and associated purchase confirmation inputs indicative of purchases thereof, the inputs being entered into an in-store consumer behavioral event metadata to a trained artificial intelligence analysis engine, the artificial intelligence analysis engine having been trained using a machine learning algorithm having an input as in-store consumer behavioral event metadata training data, and wherein the machine learning algorithm is at least in accordance with utilizing purchase confirmation data derived from the purchase confirmation inputs, the analysis engine comprising: a data interpretation controller comprising: data validation data interpretation controller and action trigger controller, action trigger controller includes: a data validation action controller, wherein the data validation data interpretation controller is configured to identify a data validation opportunity match from consumer behavioral event metadata within the store; and generating a data verification action using a data verification action controller, the data verification action configured according to the consumer behavior event metadata within the store; sending a data validation action electronic communication to at least one of the plurality of consumer electronic devices; receiving product data metadata response data from the at least one shopper electronic device in response to the data verification action electronic communication; and updating the consumable data using the product data metadata response data.
The data interpretation controller may further include: a product recommendation data interpretation controller, wherein the action triggering controller may further comprise: a product recommendation action controller, wherein, in use: the product recommendation data interpretation controller may be configured to identify product recommendation opportunities to match based on in-store consumer behavior event metadata; and the product suggestion action controller may be configured to generate and transmit a product suggestion action for display by the shopping list GUI.
The product suggestion action may include in-store location data of the product suggestion.
The in-store location data may indicate at least one of an aisle number and a shelf.
The electronic device may further comprise a sensor for sensing in-store location data, wherein the in-store consumer behavior event metadata may further comprise in-store location data.
The sensor may include at least one of a near field communication sensor and a bluetooth Beacon (BLE) sensor.
The electronic device may also include an image sensor for capturing an image of the product, wherein the product data is derivable from the image.
The product data may include at least one of a product ID and product pricing data.
The data interpretation controller may further include: the information informs the data interpretation controller; wherein, the action trigger controller may further include: a message notification action controller, wherein, in use: the information notification data interpretation controller may be configured to identify and information notification opportunity matches based on consumer behavioral event metadata within the store, and the information notification action controller may be configured to send a notification apparatus to the at least one electronic device.
The data interpretation controller may be optimized using machine learning algorithms.
The motion trigger controller may be optimized using machine learning algorithms.
The machine learning algorithm may be configured to optimize the product suggestion controller based on purchase confirmation input data derived from the shopping list GUI.
The machine learning algorithm may be configured to optimize the data validation data interpretation controller based on a probability of receiving the product data metadata response data.
The machine learning algorithm may be configured to optimize the data validation data interpretation controller based on data integrity of the product data metadata response data.
The machine learning algorithm may be configured to optimize the information controller based on consumer interactions with the information notification prompt.
The consumer product metadata may include at least one product category.
The product data metadata response data may include product price metadata.
The consumer behavior event metadata within the store may include image data, wherein the machine learning algorithm may be configured to optimize the product suggestion action controller for object recognition of the image data.
Other aspects of the invention are also disclosed.
Drawings
Although any other form is possible within the scope of the invention, a preferred embodiment of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:
FIG. 1 is a computer network for in-store consumer behavior event metadata aggregation, data verification, and artificial intelligence analysis and related action triggering thereof for data interpretation, under an embodiment;
FIG. 2 is a computing architecture illustrating in further detail servers and electronic devices of the network of FIG. 1, according to an embodiment;
FIG. 3 is an exemplary data model, controller and interface module representation of each of a server and an electronic device according to an embodiment;
FIG. 4 is a pre-inspection data flow for in-store consumer behavior event metadata aggregation, data validation, and artificial intelligence analysis and related action triggers for data interpretation, under an embodiment;
FIG. 5 is an exemplary shopping list graphical user interface displayed by an electronic device of the system of FIG. 2, according to an embodiment;
FIG. 6 is an exemplary information notification prompt displayed by an electronic device of the system of FIG. 2, according to an embodiment; and is
FIG. 7 is an exemplary personal behavior interaction with the system.
Detailed Description
Referring now to fig. 1, an exemplary network 100 of consumer electronic devices 213 is shown. It can be seen that the network 100 includes a plurality of consumer electronic devices 213, each consumer electronic device 213 being in operable communication with the analysis engine server 222 via a data network.
The consumer electronic devices 213 belong to different consumers and may be located in different locations, including those that may be classified by the network 100 according to the virtual geofence 101. in embodiments, the network 100 may represent various stores.
As will be explained in further detail below, the use of such consumer electronic devices 213 by consumers within stores enables the network 100 to receive consumer behavioral event metadata within the stores from such consumer electronic devices 213 so that the machine learning and artificial intelligence analysis described herein can be performed accordingly.
Referring now to FIG. 2, a computer system 200 is shown, according to one embodiment, the computer system 200 illustrating the analysis engine server 222 and the consumer electronic device 213 of the network 100 in further detail.
It can be seen that each of the analysis engine server 222 and the consumer electronic device 213 may take the form of a computing device having a processor 209 for processing digital data.
In an embodiment, the system 200 may utilize distributed processing of a decentralized intelligent contract blockchain platform (e.g., an etherhouse (TM) blockchain platform) as opposed to having the analytics engine server 222.
The processor 209 may be in operable communication with the memory device 223 via the system bus 208. The memory device 223 is configured to store digital data including computer program code instructions. Thus, in use, the processor 209 retrieves these computer code instructions from the memory device 223 for execution, wherein the data results may again be in-store in the memory device 223.
For ease of illustration, these computer code instructions are divided into data model 201, controller 202 and interface 203 modules.
Generally, the data model 201 includes applicable data store-internal structures (e.g., tables of a relational database) and data within the store.
In addition, the interface module 203 controls various aspects of the various user interface GUIs.
In addition, controller 202 performs various computational tasks, including interfacing with interface 203 and the data model 201 modules.
Memory device 223 may also include an operating system 207, such as a Linux kernel, or a mobile operating system that is retrieved by processor 209 during a boot phase. In embodiments using a decentralized intelligent contract blockchain platform, the memory device 223 may include all or part of the relevant blockchain ledger within the store, including data for the various transactions described herein.
Each computer device may also include an I/O interface 210 for interfacing with various computer peripherals, including in-store, sensor, and user interface peripherals.
As shown with respect to the consumer electronic device 213, the I/O interface 210 may interface with a digital display 218 for displaying digital data. The digital display 218 may overlay a tactile sensor to enable determination of user interface gestures.
I/O interface 210 may further interface with GPS sensor 216 to enable determination of the location of consumer electronic device 213, including for detecting a breach of virtual geofence 101 as described above.
In another embodiment, the consumer electronic device 213 may include other sensors 217, where the sensors may include sensors for reading product data from the consumer product (e.g., nutritional information and product appearance), aisle labels (e.g., aisle grouping of product locations and product categories within aisles), shelf labels (e.g., retailer-specific product codes/barcodes within stores, product names, shelf prices, price-specific states, product range grouping), determining location within stores, near field location technology (e.g., Near Field Communication (NFC) systems) utilizing particular aisles, and the like.
Each computer device may also include a network interface 211 for sending and receiving data over a data network 212.
In embodiments, the analytics engine server 222 is "in the cloud" and may take the form of a physical rack server, or may take the form of a virtualized server instance, such as may be implemented by Amazon Web Services (AWS). Alternatively, as described above, the analytics engine server 222 may take the form of an intelligent contract blockchain platform, where intelligent contracts are used to describe terms of peer-to-peer transactions between blockchain users without the need for a centralized server.
Further, consumer electronic device 213 may take the form of a small-sized electronic device that includes appropriate electronic circuitry for collecting and transmitting the data described herein to analysis engine server 222 over data network 212.
In an embodiment, the consumer electronic device 213 may take the form of a mobile communication device, such as a smartphone device, e.g. an apple iPhone and/or google android device, or the like. In this embodiment, to configure the mobile communication device for the particular computing process described herein, a user may download the modules 201, 202, and 203 to the memory device 223 through a downloadable software application "App," e.g., from a software application Store (e.g., Apple App Store) by the consumer electronics device 213
TM、Google Chrome App Store
TMOr Firefox App Store
TMEtc.) to download for installation and execution.
Referring now to FIG. 3, an exemplary data model 201, controller 202, and interface 203 representation of each of the analytics engine server 222 and the consumer electronic device 213 is shown.
As shown, the artificial intelligence analysis engine trained by analysis engine server 222 controller 202 may include a data interpretation controller including a consumer goods suggestion controller 308, a data verification controller 313, and an information controller 317.
The data interpretation controller module may identify various matching actions 323 from the data model 201 based on the in-store consumption behavior event metadata 324 received from the plurality of consumer electronic devices 213.
These actions 323 may then be communicated to the consumer via the interface module 203 of the analytics engine server 222 and the associated consumer electronics device 213.
In particular, the interface module 203 of the consumer electronic device 213 may display a shopping list graphical user interface 304 (shown in FIG. 5).
However, the information may be displayed on the consumer electronic device 213 in other ways (e.g., through the alert interface 310).
The data model 201 of the analysis engine server 222 may include consumer product and shopping metadata 301. In an embodiment, the consumer goods and shopping metadata 301 takes the form of a product tree that is continuously updated to account for variations between different stores and product information availability, etc., as described in further detail below.
In an embodiment, the product trees may be arranged in a hierarchy of product categories (e.g., "fast products", services (e.g., milk, bread, sugar, haircut, phone screen repair, car tuning, etc.), and among these consumer products and service categories, specific consumer products, such as dairy cow milk, paul milk, dela de cream full cream milk 1Ltr, lube oil mobile 50000km service 2013 raynaud co 2.5L4CYL four wheel drive gasoline engine MPFI 2TRA DOHC 16V (08-16), and so forth.
Product data 307 associated with products in the product tree 301 may be product data 307 representing various metadata such as product pricing, product popularity, product group data (e.g., product multi-package number, product multi-purchase partner product, product multi-purchase price, product price change partner product, nutritional information, recipes, etc.), product purchase incentives, product metadata accuracy, product metadata update incentives, product location, including in-store location and in-store location data (e.g., lane number, shelf number, etc.), and other applicable product metadata.
Further, the data model 201 may include user data 312 (e.g., demographic data, personal information, personal preferences, relationships with friends, family, user submitted metadata, etc.) representing various consumers using the system 200.
Consumer specific shopping list 316, events 324, actions 323, product data 307, product tree items 301, locations 319, and user data 312 described above with respect to what may be shopping list 316 within the user's store.
Shopping list data 316 may represent various consumer items in each consumer's shopping list (including recipe component lists, product metadata update incentive lists (e.g., price confirmation required, bar codes required), information update lists (e.g., information that may encourage an exchange for ratings of other products, etc.) and other data, such as an indication of whether the consumer purchased such items, a status of confirmation that other consumers submitted product prices from a local store, valuable insights (including consumer exchange of items on a list to another product or store, forgotten and possibly needed items not on a shopping list, location of items within a store selected by the consumer, items more capable of meeting consumer's needs than items already on a shopping list, etc.), consumer operation metadata update events, etc.
It can be seen that the data model 201 of the consumer electronic device 213 may also include corresponding manifest data 306 within the data model 201.
Thus, using the shopping list graphical user interface 304, the consumer electronic device 213 can update the list 306 using the update controller 305.
The data model 201 of the analysis engine server 222 may also include training data 321 for training a machine learning algorithm 322. In embodiments, the training data 321 may use in-store consumer behavior training data derived from the plurality of consumer electronic devices 213, and may use data from third party sources, including data from social networks (e.g., Facebook), statistical data services (e.g., the australian statistics bureau), and so forth.
As shown, the data validation and information notification controller of each product proposal, data interpretation controller may update the checklist interface 303 or generate a notification 309 displayed by an alert displayed on the shopping list graphical user interface 304 or alert interface 310 of the plurality of consumer electronic products 213.
The consumer electronic device 213 may comprise a prompt controller 311, said prompt controller 311 being configured to generate an electronic prompt in accordance with the relevant action, wherein, in an embodiment, especially for the data verification action, the information notification action and the product suggestion action, the user may respond to the product data metadata response data via a response interface 314 controlled by the response controller 318, which is then received by the receiver controller 320 of the analysis engine server 222 for updating the product tree 301 using the product tree update controller 302. Further, in embodiments, there may also be an update controller configured to update other shopping information, such as user data 312, inventory data 316, in-store data, and the like.
Referring to fig. 4, an exemplary data flow 400 for the system 200 is shown.
As shown, in-store consumption behavior event metadata 425 is received from a plurality of consumer electronic devices 213.
As described above, the in-store consumer behavior event metadata 425 may be derived, at least in part, from the shopping list graphical user interface 304 (shown in FIG. 5) displayed by the display device of each consumer electronic device 213, and may also include purchase confirmation data indicating that the consumer purchased each of the displayed consumer products.
As will be described in further detail below, shopping list graphical user interface 304 also includes a list of specific consumer items to consume and an associated purchase confirmation input (checkbox or other type of input) indicating a purchase. Thus, using such a GUI, the system is able to identify the relevant consumer items for each consumer, and further determine whether each listed item has been purchased by the consumer, metadata related to the relevant products (including availability, possible current price, quick products, etc.) in the store selected by the consumer, user metadata (including consumer segments based on the timeliness of the relevant products and shopping events and incentive thresholds for metadata collection actions related to the events), store metadata (including possible efforts to complete shopping, etc.) and event and related metadata (including metadata update incentives for products, stores, listings, users, reviews, price, availability, etc.).
Additional information may also be received from each electronic device, such as location data received via the GPS sensor 216.
In addition, other types of sensors 217 may be used to receive other information from each consumer electronic device 213.
For example, the consumer electronic device 213 may include an image sensor, e.g., to allow a consumer to capture image data of a product, to allow the system 200 to identify the product, e.g., based on various visual characteristics of the product, including reading a bar code provided thereon (including reading an image of the product into a trained artificial intelligence object recognition engine).
Alternatively, pricing data may be captured, as opposed to capturing image data of the product, such as by capturing an image of the adjacent product displayed by the price label using sensor 217.
Alternatively, rather than capturing image data of the product, activity level, aesthetics, comfort, etc. may be captured, such as by capturing an image of the front of the store, a nearby store, or a parking lot using the sensor 217.
Alternatively, instead of capturing image data of the product, proof of purchase, in-store attendance, or the like may be captured, for example, by capturing an image of the two-dimensional code using the sensor 217.
Alternatively, as opposed to capturing image data of a product, pricing information (price, unit price, special price, discount rate, product, proof of purchase), inventory information (including purchased product, in-store metadata, total price of purchase, method of purchase, registration number, registrar details, in-store manager details, purchase amount, product receipt alias, tax status, etc.) may be captured, such as by capturing an image of a purchase receipt using sensor 217.
In a further embodiment, the sensor 217 is configured to determine the location of the consumer within the store, specifically the resolution down to the aisle (including aisle number, aisle grouping and nearby products, etc.) and, in a further embodiment, shelf resolution.
For example, within a store, a plurality of near field communication tags may be provided in various aisles or shelves, which are then scanned by the sensor 217 to determine the location of the consumer electronic device 213 within the store.
Alternatively, bluetooth beacon technology may be used, where the sensor 217 utilizes received signal strength measurements to determine the location of the consumer electronic device 213 relative to one or more bluetooth beacons within the store.
In an embodiment, and as suggested in FIG. 5, such in-store location data may be derived from user input, including for consumer electronic devices 213 that do not have such sensors 217.
As shown, event metadata 402 is from in-store consumer behavior event metadata 425.
Such event metadata 402 can represent various consumer events, such as purchasing a product, entering a store (as determined by an inward breach of virtual geo-fence 101), returning to a consumer's home (as determined by an outward breach of virtual geo-fence 101), quickly selecting a product, product optimization, going home (an "home" location geo-fence assumed by an internal breach), accepting an incentive (such as a metadata update, an information incentive, etc.), a redemption incentive, and other consumer activities.
As shown, event metadata 402 may include event type metadata 403 indicating a consumer behavior event type and location metadata 404 indicating a location of the event.
Consumer behavioral events and location metadata 404 may be derived from GPS sensor 216 and, in embodiments, may be used by system 200 to determine which store a consumer is currently in, and the important locations of events outside the store (including home, work, commute start and end, etc.). The location of the significant events may be determined by an artificial intelligence model that is trained to identify the most desirable locations (notifications, suggestions, incentives, etc.). The model takes into account not only the current user's location, but also other locations and attributes of other users in the network at that time.
The event metadata 402 may also include higher resolution in-store location data 405, the in-store location data 405 representing the user's location within the store, such as particular aisles, shelves, and the like.
Event metadata 402 may also include manifest metadata 407, which manifest metadata 407 may include a consumer-specific shopping list including associated metadata including metadata indicating that the consumer purchased his or her shopping.
In addition, the event metadata 402 may also include product metadata 408, the product metadata 408 being metadata applicable to various consumer products. Such product metadata 408 may represent product price, product location, product availability, or other applicable product metadata.
As shown, the event metadata 402 is fed into a trained artificial intelligence analysis engine 413.
As input such data, the trained artificial intelligence analysis engine 413 is configured to generate intelligent artificial intelligence insights 423, which can then be communicated back to the relevant electronic devices of the network 213. Such insights refer to any intelligent data and/or notifications that may be useful to a consumer network during shopping.
In particular, the trained artificial intelligence analysis engine 413 may include a data interpretation controller 414. The data interpretation controller 414 interprets the event metadata 402 so that various opportunities for generating related actions can be matched.
Specifically, as shown, the trained artificial intelligence analysis engine 413 can include an action trigger controller 418.
Thus, for any potential matching opportunities identified by the data interpretation controller 414 using the aggregated event metadata 402, the trained artificial intelligence analysis engine 413 can trigger associated and applicable actions using the action trigger controller 418.
As indicated above, and as described above, the data interpretation controller 414 may include a product suggestion controller 308, the product suggestion controller 308 configured to intelligently suggest various consumer products. Such consumer products may be suggested based on consumer-specific parameters (including demographics, previous purchasing behavior) and other consumer-specific parameters (such as consumption behavior habits of other consumers).
Data interpretation controller 414 may also include a data verification controller 313, data verification controller 313 configured to identify opportunities for enhancing the integrity of consumable metadata 301, inventory metadata, in-store metadata, and the like.
For example, if the price of a particular consumer product is "soft" (i.e., identified as potentially unreliable), data validation controller 313 can identify to generate an associated data validation action from data validation action controller 420, which data validation action controller 420 can be communicated to an associated electronic device network device 213 for validation by the consumer, e.g., to validate price by the consumer, to add a barcode, to validate aisle, to validate how busy a store is, to validate product quality, etc.
The data interpretation controller 414 may also include an information controller 317, the information controller 317 being configured to generate various information notification actions from an information notification action controller 421, the information notification action controller 421 typically being transferable in a push manner to associated electronic device notifications of the network 213, other GUI alerts (e.g., newsfeed items, upgrade items on a comparison screen), e-mails, etc.
As shown in data flow 400, such crowd-sourced aggregated event metadata 402 may be fed into machine learning algorithm 322.
In an embodiment, the machine learning algorithm 322 may include an optimizer configured to optimize each module of the data interpretation controller 414. Specifically, as shown, the machine learning algorithm 322 may include a product recommendation model optimizer 410 for generating training data 422 for optimizing the product recommendation controller 308. Similarly, the machine learning algorithm 322 may include a data validation model optimizer 411 and an information notification model optimizer 412 for optimizing the respective data validation controller 313 and information controller 317.
Product suggestion model optimizer 410 can be configured to optimize the purchase of suggested products. The data validation model optimizer 411 may be configured to optimize the accuracy of the network-wide metadata and the likelihood or probability of receiving consumer feedback, or alternatively, the integrity of receiving feedback.
Information notification model optimizer 412 may be configured to optimize notifications according to the acceptance of such notifications, for example, acceptance of notification 601 as shown in FIG. 6.
For example, the machine learning algorithm 322 may train the aggregated event metadata 402 using a Recurrent Neural Network (RNN) model, and may identify that a particular product suggestion is purchased by a consumer more frequently than another product suggestion (determined from the shopping list graphical user interface 304), and thus may bias the product suggestions accordingly.
It should be noted that the machine learning algorithm 322 may be further configured to optimize the action trigger controller 418 in order to optimize the triggering of the associated action.
In one example, the machine learning algorithm 322 can determine that if the product suggestion is made before the consumer electronic device 213 enters the virtual geofence 101 of the applicable store, the product suggestion is more likely to be purchased by the consumer, and thus the action trigger controller 418 can be optimized accordingly.
As another example, the machine learning algorithm 322 can recognize that if a woman is suggested to purchase a product before the virtual geofence 101 breaks inward, then the woman is more likely to purchase the product, while a man is more likely to purchase the product only at the time of the suggestion in the store (i.e., once the consumer electronic device 213 has violated the virtual geofence 101 of the applicable store), so the action trigger controller 418 can be optimized accordingly.
In an embodiment, the trained artificial intelligence analysis engine 413 may take the form of an artificial neural network, and thus the trained data 422 may represent the optimized weights for each node of the neural network.
Referring now to FIG. 5, an exemplary graphical user interface 500 displayed by the consumer electronic device 213 is shown.
As shown, the shopping list includes a variety of snack products, such as milk, pasta and bread, in addition to a special product, namely a highly sensitive toothpaste.
A user may configure such a manifest by entering such items. In one embodiment, as a user type, interface 500 suggests products from product tree 301 using text prediction. In further embodiments, the manifest may populate previously purchased, frequently purchased, or predicted consumer products that the consumer needs using AI insights (as suggested actions from previous events, by the user or even other users in the network). In further embodiments, the text of the web page data (whole web page or text selected by the user, etc.) may be parsed from the product reference list, or may be intelligently analyzed to override controls on the interface 500 (including for product suggestions 501 (product addition, product comparison, product search, and new product examples), and data verification prompts 503 (submitting product reviews (including sending to friends on the network) as well.
Now, as shown, interface 500 includes a product recommendation 501 generated by product recommendation controller 308 and product recommendation action controller 419, and a data validation prompt 503 generated by data validation controller 313 and data validation action controller 420.
Specifically, for milk instant products, the shopping list interface 501 generates product recommendations, where, for example, specific and intelligent recommendation products in this instant product category are recommended, such as dairy cow milk and paul milk.
Additional artificial intelligence insights may be relevant to this, including location and pricing insights, where, for example, it can be seen that for recommended dairy cow milk, a particular location of the store (third channel and second shelf from the top) is displayed, for paul milk, which is in fact the cheapest in the store at the present time.
However, as shown, due to the lower price, the product recommendations may suggest placing the product in another location. As shown, the product recommendations 501 also include recommendations for Kells milk that requires only $1, but requires walking 250 meters to another location.
For pasta quick products as shown, the product suggestions 501 may further intelligently recommend a particular type of pasta. As described above, suggested pasta types may be generated in an intelligent manner by the trained artificial intelligence analysis engine 413 for optimization for purchase by consumers. For example, a particular type of consumer product may be from consumer-specific metadata, e.g., where gluten-free pasta is suggested for a particular consumer.
Alternatively, suggested pasta of a particular type may be suggested based on analysis of event metadata 402 received from other consumers (including similar consumers and other parameters).
As shown, the product suggestions 501 may also intelligently suggest related items, such as pasta sauce, where the machine learning algorithm 322 considers that consumers often purchase these types of products together, which may belong to different product categories.
Using the trained artificial intelligence analysis engine 413 may further generate non-intuitive artificial intelligence insights in which, for example, for bread loaf products, the interface 500 may suggest shoe polishing. Although this may not seem intuitive, in practice the shoe polish may be very relevant to the user.
As shown, interface 500 displays data verification prompt 503, where data verification controller 313 is identified as an opportunity to enhance the integrity of product data 307, as described above.
As shown, for the collyrium sensitive toothpaste, the data validation prompt 503 asks the user if the price is actually $ 2.99, and furthermore, if it is actually on lane 3. As shown, if so, the user can check the check box or enter the correct amount.
In a further embodiment, the system 200 may be configured to perform OCR analysis on the consumer purchase receipt to enable appropriate product data 307, including price data, to be collected accordingly.
As shown, the interface 500 includes an OCR receipt button 504, which button 504 may be used by a user when capturing an image of a purchase receipt using the image sensor 217 of the consumer electronic device 213.
For each consumer product on the shopping list interface 500, the user can use the associated purchase confirmation input (check box, etc.) to indicate whether the product has been purchased.
Thus, by utilizing such purchasing behavior feedback from the purchase confirmation input, the machine learning algorithm 322 is also able to train the artificial intelligence analysis engine 413 accordingly.
Using the trained artificial intelligence analysis engine 413 may further generate anti-fraud artificial intelligence insights in which, for example, a user may maliciously submit an incorrect price update or fraudulent photo proofs of their purchases to reward a brand.
The trained artificial intelligence analysis engine 413 may be configured to identify anomalies by using a class of Support Vector Machines (SVMs) or a class of neural network models. The data verification controller 313 may not only be related to purchasing behavior, but may also be related to pre-purchase behavior, such as when and where items were added to the inventory, and the user's past history of such updates and claims, the data verification controller 313 may identify abnormal and irregular behavior. This will generate an associated data verification action from the data verification action controller 420, which the data verification action controller 420 may be transmitted to the user's electronic device to ask the user to further prove that the item was purchased (e.g., scan a barcode at home).
For example, a brand may wish to know the current in-store status of its products on a particular local supermarket shelf, aisle to ensure that its products are correctly located, or that the inventory on the shelf is sufficient and correct. When a user is shopping at a local supermarket, the data verification controller 313 can predict from the user's purchase confirmation input (e.g., input through a check-box user interface, etc.) for other items in the store and other current application behaviors (e.g., breaking the geofence perimeter inward), that the user can be in the same aisle as the brand of product for which current in-store condition information is desired. Thus, the action trigger controller 418 may generate an associated data verification action using the data verification action controller 420, which the data verification action controller 420 may be transmitted to the user's electronic device to prompt the user to take a picture of the shelf for the brand, perhaps for the reward.
For example, the user may submit a new product description. The trained artificial intelligence analysis engine 413 may be configured to generate intelligent artificial intelligence classes for "fast products," for example, by using multiple classes of decision forests or multiple classes of neural network models. This will generate a relevant data verification action from the data verification action controller 420, which data verification action controller 420 may then be transmitted to the relevant electronic devices of the network 213, so that these users can verify and verify the suggested classification
Referring now to fig. 6, an information notification 601 generated by the information controller 317 is shown.
As can be seen, the notification 601 has determined the location of the consumer electronic device 213, alerts the user to the proximity of the smart product suggestion, and further displays information regarding the determined price of the suggested product.
By using a button input, the user can activate a navigation interface to navigate to the product, or cancel the alert.
Various exemplary embodiments will now be provided to further illustrate the functionality of the system 200. It should be noted that these embodiments are merely exemplary, and any technical limitation must not be correspondingly attributed to all embodiments.
As a first example, a user selects a quick product using shopping list GUI 500.
In response, the product suggestion controller 308 intelligently suggests particular products that are related to the quick product. Further, if the data validation controller 313 identifies any product metadata associated with the fast product as unreliable ("soft"), the data validation controller 313 may trigger a data validation action from the data validation action controller 420 seeking to collect further product metadata accordingly.
Further, the information controller 317 may identify information related to the quick product and notification action accordingly.
As described above, the machine learning algorithm 322 may train the data interpretation controller 414 and/or the action trigger controller 418 to optimize the likelihood that the consumer accepts or otherwise interacts with the suggested product, data validation requests, and/or informational notifications.
In another example, the system 200 determines that the consumer electronic device 213 has breached the perimeter of the virtual geofence 101 surrounding the store inward by determining the location of the consumer electronic device 213. Thus, the system 200 can infer that the user is likely to shop.
Upon identifying such a possible event, data interpretation controller 414 may implement any potentially matching actions from action trigger controller 418.
Further, various related data may be updated in response to such events, e.g., the system 200 can retrieve product data, product specials, etc. from within the appropriate store. Further, the system 200 may analyze recent or substantial real-time event metadata 402 associated with other consumers within the same area. Thus, data interpretation controller 414 may operate using these updated information.
In another embodiment, the user confirms the purchase or selection of a product (e.g., by lifting the product from the store and placing it in a cart or basket) using a purchase confirmation input (such as the check box input shown in FIG. 5).
Thus, from the particular product indicated, the system 200 can infer other information, such as the in-store location of the consumer, such as the resolution of the present walkway. Thus, the product recommendation controller 308 is able to recommend other products, such as other products in the same or adjacent aisles, at the same in-store location.
Further, upon receiving confirmation of the shipment of such products, the product suggestion controller 308 may therefore not recommend the same or similar products.
In addition, shopping list interface 500 may suggest that the next product be selected based on proximity.
Other information may be gathered from the consumer's interaction with shopping list GUI 500, such as the time it takes to complete a store, such as a timestamp between the first and last item text from shopping list interface 500.
Further, the frequency or number of interactions of the consumer using the interface 500 may be utilized to infer how busy a particular store is in order to allow the trained artificial intelligence analysis engine 413 to possibly suggest shopping at another time or place. Such information may also be collected from the user through an information notification prompt that asks the user for current busy status in the store or other in-store information.
In a further example, using the data validation prompt 503, the user submits a price for the in-store consumable item.
Accordingly, data verification controller 313 may determine that the consumer is willing to provide such feedback, and thus may increase the frequency of subsequent requests.
Further, in this case, for a particular product, the data verification controller 313 may request information related to the related consumer product, such as other consumer products within the same channel.
In another example, the system 200 detects an external breach of the virtual geofence 101 by the consumer electronics device 213 indicating that the consumer has left a location within the store. Thus, the system 200 can infer that the user has completed shopping.
Thus, in an embodiment, the product suggestion controller 308 may suggest consumer products that may be desired by the consumer at other locations near the store location, such as recommending newspapers from neighboring news agencies.
Further, event metadata 402 may be analyzed to identify correlations between consumers and store locations, such as by analyzing the number of consumers within parameters and the time it takes to complete a shopping trip to estimate the ease of parking.
In another example, the system 200 can detect an inward breach of the home virtual geo-fence 101 indicating that the user has returned to the home. Thus, the system 200 can infer that the user is likely to unpack the purchased consumable.
Thus, during the unpacking process, the data verification controller 313 may prompt the user for a data verification action from the data verification action controller 420 to provide metadata related to the various products purchased.
In addition, the data validation controller 313 can receive OCR data on the purchase receipt to be able to determine product metadata therefrom, such as by keyword matching line items and identifying relevant pricing information (price, unit price, special price, discount rate, product, proof of purchase), inventory information (including purchased product, store metadata, total price of purchase, purchase mode, registration number, registrar details, store manager details, purchase amount, purchase quantity, product receipt alias, tax payment status, etc.)
Further, the data validation controller 313 can receive OCR data of shopping-related documents (including product packaging labels, aisle labels, shelf labels/tickets, etc.) to enable shopping metadata (including products, aisles, shelves, etc.) to be determined therefrom, such as by keyword matching characters and identifying related shopping information (including nutritional information, ratings, product location, aisle grouping of product categories within aisles, retailer-specific product codes/barcodes within stores, product name, shelf price, price-specific status, product range grouping, etc.).
Further, for any consumer product having an associated store that specifically requires purchasing the product, the information controller 317 may prompt the user for an information notification operation from the information notification operation controller 421 for the product, where, for example, the prompt may notify the user that the user is eligible to apply for a credit award for future shopping trips if the user is to take an image of a particular product.
Referring now to FIG. 7, an exemplary user scenario 700 illustrating the features and functionality of system 200 is shown. It should be noted that scenario 700 is merely exemplary and that technical limitations need not be attributed to all embodiments of system 200 accordingly. Further, in the exemplary scenario 700, the in-store consumer behavior event metadata 425 is displayed as a circle, the data verification actions from the data verification action controller 420 are displayed as a parallelogram, the information notification actions from the information notification action controller 421 are displayed as a rectangle, the product suggestion operations from the product suggestion action controller 419 are represented as an ellipsis, and the data updates to the data model 201 are displayed as a hexagon.
Scenario 700 now begins at step 712, where user a adds "HP show 120 g" to the shopping list of local store 1 at home, without a price or barcode. Thereafter, event 712 may cause system 200 to trigger a series of data validation actions from data validation controller 313, which data validation controller 313 includes a first action 713 in which system 200 sends data validation request 713 to other consumer electronic devices 213 to request a price from the user who recently purchased the same product. Alternatively, or in addition, action 714 may be sent to other consumer electronic devices 213 to request barcodes from other users who have recently purchased the same product.
Upon receiving data validation actions 713, 714, user B's electronic device may generate an event 716 where user B updates the price of the item using a photograph of the price entry on his shopping receipt while at home. Thus, event 716 may further cause system 200 to perform a data update 717 to update the price of the item using the provided pricing information, but to set a flag indicating that the price is "unconfirmed". Further, system 200 may then initiate data validation act 718 to generate a data validation act to request pricing confirmation from other electronic devices of other users, e.g., to confirm that photos of price items on their shopping receipts match products and prices from data validation act 713.
Further, in the event that user a has performed a product addition event 712 (by typing in a search term that matches the product, by capturing an image of a product package, by scanning a barcode, etc.), the system 200 may generate an AI-generated product suggestion action 711, e.g., suggesting that user a add the supplemental item category "oven baked chips" to a shopping list from which the system 200 has learned to use the machine learning algorithm 322 typically associated with "HP Sauce120 g".
Event 712 may further cause system 200 to initiate price request data validation action 715 for consumer electronic device 213 of user C, because system 200 has determined that user C may be within store 1 or within the same aisle in the near future. Such predictions may be generated by machine learning algorithms 322, with machine learning algorithms 322 training on various event metadata 402 (including consumer behavioral events and location metadata 404).
Upon receiving price request data validation action 715, user C's electronic device may generate event 701, where user C updates the price in-store zone 1 from, for example, $1 to $1.30, and captures an image of a shelf price label or ticket. In response, the system 200 performs a data update 702 to update prices in the stores and, if applicable, corresponding prices for related stores of the same retailer (price reconciliation).
Further, the data verification controller 313 may generate a data verification action from the data verification action controller 420 and send a verification request 703 to the consumer electronic device 213 of the other user D to confirm the price. When user D confirms that the shelf price matches the picture taken, the system 200 performs a price confirmation update 705.
Further, in response to event 701, system 200 may initiate product suggestion 706 to notify all other users of the system that there is a "brown sauce" (i.e., item category) on their associated shopping list, e.g., 120g HP sauce sold at a special price of $ 1.30. In this regard, the information notification action 707 may be sent to other electronic devices.
The system 200 may further generate another product suggestion 708 to notify all other users with "HP Sauce120 g" on their shopping list that the item is premium $1.30 at store number 1. It should be noted that by user C updating the price at event 701, the system 200 can notify the original user of the new price for the item originally added by the original user by checking the relevant metadata, even if the original user did not specifically require a price update.
Price update event 701 can also include a send information notification action 709, the notification action 709 notifying a user who can shop at store 1, that there is at least one other user currently active in the store (for an occurrence such as a price check request, other users may be interested, availability request, etc.). When user E receives the information notification action 709, the electronic device 230 may generate an event 719 in which user E wishes to know the price of hugging diapers in store 1 and provide the point of transaction user C50 to ascertain. In response, the system 200 may generate a data verification action 720 to send a notification to the consumer electronic device 213 of user C.
Event 701 may further trigger another product suggestion 710, which suggestion 710 informs a user that has "HP Sauce120 g" on the shopping list for store number 2, which may be a premium of $1.30 given that store number 2 is operated by the same retailer.
The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art that the specific details are not required in order to practice the invention. Thus, the foregoing descriptions of specific embodiments of the present invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed; obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. The following claims and their equivalents define the scope of the invention.
