阅读说明：本技术 文书评审的校对方法及装置、存储介质、电子设备 (Document review proofreading method and device, storage medium and electronic equipment ) 是由 郑立颖 徐亮 阮晓雯 于 2019-09-18 设计创作，主要内容包括：本公开提供了一种文书评审的校对方法及装置,属于相似度匹配技术领域,该方法包括：获取待校对文书,并将所述待校对文书按照预定规则进行拆分,以得到校对模板；获取针对所述待校对文书的条款目录中条款的标注信息,并将所述标注信息按照所述预定规则进行拆分,以得到待校对文本；将所述校对模板和与其相对应的所述待校对文本进行对比；当所述校对模板和与其相对应的所述待校对文本的内容一致时,则确定所述待校对文本为正确标注。该方法提高了文书评审校对的效率,降低了校对成本。(The utility model provides a proofreading method and a device for document review, which belongs to the technical field of similarity matching, and the method comprises the following steps: acquiring a document to be collated, and splitting the document to be collated according to a preset rule to obtain a collation template; acquiring the labeling information of the clauses in the clause catalog of the document to be corrected, and splitting the labeling information according to the preset rule to obtain a text to be corrected; comparing the proofreading template with the text to be proofread corresponding to the proofreading template; and when the contents of the proofreading template and the text to be proofread corresponding to the proofreading template are consistent, determining that the text to be proofread is correctly marked. The method improves the efficiency of document review and proofreading and reduces the proofreading cost.)

1. A proofreading method for document review is characterized by comprising the following steps:

acquiring a document to be collated, and splitting the document to be collated according to a preset rule to obtain a collation template;

acquiring the labeling information of the clauses in the clause catalog of the document to be corrected, and splitting the labeling information according to the preset rule to obtain a text to be corrected;

comparing the proofreading template with the text to be proofread corresponding to the proofreading template;

and when the contents of the proofreading template and the text to be proofread corresponding to the proofreading template are consistent, determining that the text to be proofread is correctly marked.

2. The proofreading method of document review according to claim 1, wherein comparing the proofreading template with the text to be proofread corresponding thereto comprises:

obtaining statement sentence vectors of the proofreading template;

obtaining a sentence vector of the text to be proofread corresponding to the proofreading template;

calculating the cosine similarity between the statement sentence vector of the proofreading template and the statement sentence vector of the text to be proofread corresponding to the statement sentence vector;

and determining whether the contents of the proofreading template and the text to be proofread corresponding to the proofreading template are consistent or not according to the cosine similarity.

3. The proofreading method of document review according to claim 2, wherein determining whether the proofreading template and the content of the text to be proofread corresponding thereto are consistent according to the cosine similarity comprises:

and when the cosine similarity is larger than or equal to a preset threshold value, determining that the proofreading template is consistent with the content of the text to be proofread corresponding to the proofreading template.

4. The proof reading method for document review according to claim 3, wherein the determination of the predetermined threshold value comprises:

obtaining statement sentence vectors of samples of the proofreading template, wherein the samples are preset synonymous statements of the proofreading template;

selecting a test sample from the samples, and calculating cosine similarity of sentence vectors between the test sample and other samples;

and determining the predetermined threshold value based on cosine similarity of sentence vectors between the test sample and other samples.

5. The proofreading method of document review according to claim 1, further comprising, after comparing the proofreading template with the text to be proofread corresponding thereto:

when the contents of the proofreading template and the text to be proofread corresponding to the proofreading template are inconsistent, performing character matching on the proofreading template and the text to be proofread corresponding to the proofreading template;

and when the coincidence rate obtained by character matching between the proofreading template and the text to be proofread corresponding to the proofreading template reaches a preset proportion, determining that the text to be proofread is the correct label.

6. The document review proofreading method according to claim 5, wherein after determining that the text to be proofread is correctly labeled when a coincidence rate obtained by character matching between the proofreading template and the text to be proofread corresponding thereto reaches a predetermined ratio, the method further comprises:

and replacing the content of the text to be proofread with the content of the proofreading template corresponding to the content of the text to be proofread.

7. The method of claim 1, wherein after the document to be collated is split according to the predetermined rule to obtain the collation template, the method further comprises:

acquiring the position number of the proofreading template in the document to be proofread;

wherein, the comparing the proofreading template with the text to be proofread corresponding to the proofreading template comprises:

and comparing the text to be proofread with the proofreading template, wherein the position numbers of the text to be proofread are the same.

8. A proofreading apparatus for document review, comprising:

the system comprises a first acquisition module, a second acquisition module and a verification module, wherein the first acquisition module is used for acquiring a document to be verified and splitting the document to be verified according to a preset rule to obtain a verification template;

the second acquisition module is used for acquiring the labeling information of the clauses in the clause catalog of the document to be corrected and splitting the labeling information according to the preset rule to obtain a text to be corrected;

the comparison module is used for comparing the proofreading template with the text to be proofread corresponding to the proofreading template;

and the determining module is used for determining that the content of the text to be collated is correctly marked when the collation template is consistent with the content of the text to be collated corresponding to the collation template.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a proof reading method of a document review according to any one of claims 1 to 7.

10. An electronic device, comprising:

a processor; and

a memory having a computer program stored thereon;

wherein the processor is configured to implement the proofreading method of the document review of any one of claims 1-7 via execution of the computer program.