CN113240556A

CN113240556A - Infringement processing method, device, equipment and medium based on intelligent decision

Info

Publication number: CN113240556A
Application number: CN202110601871.9A
Authority: CN
Inventors: 张师琲
Original assignee: Ping An Technology Shenzhen Co Ltd
Current assignee: Ping An Technology Shenzhen Co Ltd
Priority date: 2021-05-31
Filing date: 2021-05-31
Publication date: 2021-08-10
Anticipated expiration: 2041-05-31
Also published as: CN113240556B

Abstract

The embodiment of the application provides an infringement processing method, an infringement processing device, infringement processing equipment and an infringement processing medium based on intelligent decision, which are applied to the technical field of artificial intelligence, wherein the method comprises the following steps: extracting first key information from the first enterprise information and extracting second key information from the second enterprise information; comparing the first key information with the second key information to obtain a comparison result, and determining a preliminary infringement judgment result according to the comparison result; searching for article segments related to a first business and related to a second business; carrying out sentiment classification on the article fragments and carrying out article property classification to obtain a classification result; and generating right-maintaining reference information according to the preliminary infringement judgment result and the classification result. By adopting the method and the device, the illegal competition and the infringement behavior can be found in time, and the data referential performance of the right maintaining process aiming at the illegal competition and the infringement behavior is improved. This application relates to blockchain techniques, such as where entitlement reference information may be written into a blockchain.

Description

Infringement processing method, device, equipment and medium based on intelligent decision

Technical Field

The present application relates to the field of intelligent decision-making technologies, and in particular, to an infringement processing method, apparatus, device, and medium based on intelligent decision-making.

Background

In the intellectual property system, one right-maintaining item is called illegal competition, which means that operators and other related market participants adopt accepted commercial methods of violating fairness, honesty credit and the like to compete for transaction opportunities or destroy the competitive advantages of others, the legal rights and interests of consumers and other operators are damaged, and the social and economic order is disturbed.

Now, the internet technology is developed, and many illegal competitions and infringement behaviors on the network are difficult to maintain if not discovered or lifted by people. The identification of fraudulent competition and infringement is mainly performed by simple image matching or text matching, and this is also generally performed after the occurrence of fraudulent competition and infringement. It can be seen that the above process makes the discovery of illicit competition and infringement less timely. Also, the result obtained by such simple image matching or text matching is less referable to data of the right maintaining process for illegal competition and infringement acts.

Disclosure of Invention

The embodiment of the application provides an infringement processing method, device, equipment and medium based on intelligent decision, which can timely discover illegal competition and infringement behaviors and can improve data referential performance of a right maintenance process aiming at the illegal competition and the infringement behaviors through right maintenance judgment information.

In a first aspect, an embodiment of the present application provides an infringement processing method based on an intelligent decision, including:

acquiring first enterprise information of a first enterprise and second enterprise information of a second enterprise;

extracting first key information for infringement judgment from the first enterprise information, and extracting second key information for infringement judgment from the second enterprise information;

comparing the first key information with the second key information to obtain a comparison result, and determining a preliminary infringement judgment result according to the comparison result;

searching for article segments related to a first business and related to a second business;

carrying out sentiment classification on the article fragments and carrying out article property classification to obtain a classification result, wherein the classification result comprises a sentiment class and an article property class;

and generating right-maintaining reference information according to the preliminary infringement judgment result and the classification result.

Optionally, the first enterprise information includes a trademark of a first enterprise, and the extracting first key information for infringement determination from the first enterprise information includes:

acquiring text content included by the trademark of the first enterprise;

calling a word filtering model to identify common words appearing in the text content, and filtering the common words appearing in the text content to obtain keywords of the trademark of the first enterprise; the word filtering model comprises filtering rules set according to a general word bank;

and determining the keyword of the trademark of the first enterprise as first key information.

Optionally, the first enterprise information includes an enterprise name of the first enterprise, and the extracting first key information for infringement determination from the first enterprise information includes:

calling a word size extraction model to extract an enterprise word size included by the enterprise name of the first enterprise; the font size extraction model comprises extraction rules set according to the font size of the enterprise;

and determining the enterprise word number included by the enterprise name of the first enterprise as first key information.

Optionally, the acquiring second enterprise information of the second enterprise includes:

filtering an enterprise information base according to target enterprise information of a first enterprise and a filtering rule set based on the enterprise information to obtain a first enterprise information set, wherein the enterprise information base comprises a plurality of enterprise information, and the first enterprise information set comprises at least one enterprise information with the first character same as that of the target enterprise information;

determining the operation range of the first enterprise according to the target enterprise information;

and filtering the first enterprise information set according to the operation range of the first enterprise and a filtering rule set based on the operation range to obtain a second enterprise information set, wherein the second enterprise information set comprises enterprise information of a second enterprise with the operation range consistent with that of the first enterprise.

Optionally, the method further includes:

carrying out named entity recognition on the text fragment to obtain a target named entity;

generating right-maintaining reference information according to the preliminary infringement judgment result and the classification result, wherein the right-maintaining reference information comprises:

and generating right-maintaining reference information according to the preliminary infringement judgment result, the target named entity and the classification result.

Optionally, the method further includes:

extracting a target triple from the article fragment, wherein the target triple is composed of an entity of a first enterprise, an entity of a second enterprise and a first incidence relation between the first enterprise and the second enterprise;

acquiring a authenticity verification result of the first association relation according to the enterprise knowledge graph and the target triple;

and generating right-maintaining reference information according to the preliminary infringement judgment result, the classification result and the authenticity verification result of the first association relation.

Optionally, the method further includes:

extracting target event information from the article fragment;

extracting a set of event information from the enterprise knowledge graph, the set of event information comprising first event information associated with the first enterprise and second event information associated with the second enterprise;

inquiring whether the event information set comprises the target event information or not, and obtaining a true and false verification result of the target event information according to an inquiry result;

and generating right-maintaining reference information according to the preliminary infringement judgment result, the classification result and the authenticity verification result of the target event information.

In a second aspect, an embodiment of the present application provides an infringement processing apparatus based on an intelligent decision, including:

the acquisition module is used for acquiring first enterprise information of a first enterprise and second enterprise information of a second enterprise;

the extraction module is used for extracting first key information for infringement judgment from the first enterprise information and extracting second key information for infringement judgment from the second enterprise information;

the comparison module is used for comparing the first key information with the second key information to obtain a comparison result, and determining a preliminary infringement judgment result according to the comparison result;

a search module for searching article segments related to a first enterprise and related to a second enterprise;

the classification module is used for carrying out sentiment classification on the article fragments and carrying out article property classification to obtain a classification result, and the classification result comprises a sentiment type and an article property type;

and the information generation module is used for generating right-maintaining reference information according to the preliminary infringement judgment result and the classification result.

In a third aspect, an embodiment of the present application provides a computer device, including a processor and a memory, where the processor and the memory are connected to each other, where the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method according to the first aspect.

In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, where the computer program is executed by a processor to implement the method according to the first aspect.

In summary, the computer device may obtain first enterprise information of a first enterprise and second enterprise information of a second enterprise, extract first key information for infringement determination from the first enterprise information, and extract second key information for infringement determination from the second enterprise information; the computer equipment compares the first key information with the second key information to obtain a comparison result, and determines a preliminary infringement judgment result according to the comparison result; the computer equipment searches article fragments related to a first enterprise and a second enterprise, and carries out emotion classification on the article fragments and article property classification to obtain classification results, wherein the classification results comprise emotion categories and article property categories; the computer equipment generates right-maintaining reference information according to the preliminary infringement judgment result and the classification result; compared with the prior art that simple text matching or image matching operation is executed after the illegal competition and the infringement behavior appear, the method and the device can find the illegal competition and the infringement behavior in time so as to maintain the right actively, and the data referential property of the right maintaining process aiming at the illegal competition and the infringement behavior can be improved by the right maintaining judgment information obtained by the method and the device.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic flowchart of an infringement processing method based on intelligent decision according to an embodiment of the present disclosure;

FIG. 2 is a schematic flow chart illustrating an intelligent decision-based infringement processing method according to yet another embodiment of the present application;

fig. 3 is a schematic structural diagram of an infringement processing apparatus based on intelligent decision according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

Please refer to fig. 1, which is a flowchart illustrating an infringement processing method based on intelligent decision according to an embodiment of the present application. The method may be applied to a computer device. The computer device may be a server or an intelligent terminal. The server can be an independent server or a server cluster, and the intelligent terminal can be a desktop computer, a notebook computer and other intelligent terminals. Specifically, the method comprises the following steps:

s101, first enterprise information of a first enterprise and second enterprise information of a second enterprise are obtained.

Wherein the first enterprise information may include at least one of: a trademark of the first enterprise, an enterprise name of the first enterprise, an enterprise abbreviation of the first enterprise, and an enterprise font size of the first enterprise. The first enterprise may be a rights issuer. The second enterprise information may include at least one of: a trademark of the second enterprise, an enterprise name of the second enterprise, an enterprise abbreviation of the second enterprise, and an enterprise font size of the second enterprise. The second enterprise may be a suspected infringer.

In one embodiment, the computer device may receive the enterprise information of the first enterprise and the enterprise information of the second enterprise sent by the terminal device. In an application scenario, the computer device may obtain an enterprise information interface through the terminal device, a terminal user may input enterprise information of a first enterprise and enterprise information of a second enterprise based on the enterprise information interface, and the terminal device may obtain the enterprise information of the first enterprise and the enterprise information of the second enterprise, and send the enterprise information of the first enterprise and the enterprise information of the second enterprise to the computer device after clicking the confirmation button.

S102, extracting first key information for infringement judgment from the first enterprise information, and extracting second key information for infringement judgment from the second enterprise information.

S103, comparing the first key information with the second key information to obtain a comparison result, and determining a preliminary infringement judgment result according to the comparison result.

In steps S102-S103, the computer device may extract first key information for infringement determination from the first business information and extract second key information for infringement determination from the second business information. The computer equipment can obtain a comparison result by comparing the first key information with the second key information, and can determine a preliminary infringement judgment result according to the comparison result. The first key information is key information extracted from the first enterprise information and used for infringement judgment. And the second key information is key information extracted from the second enterprise information and used for infringement judgment. The preliminary infringement determination result may be a result indicating whether the second business information infringes the first business information.

Several ways of extracting the first key information for infringement determination from the first business information are explained below.

In one embodiment, when the first business information includes a trademark of the first business, the computer device extracts the first key information for infringement determination from the first business information by: the computer equipment acquires text content included by the trademark of the first enterprise, calls a word filtering model to identify common words appearing in the text content, and filters the common words appearing in the text content to obtain keywords of the trademark of the first enterprise, so that the keywords of the trademark of the first enterprise are used as first key information. That is, the first key information may be a keyword of a trademark of the first enterprise obtained through the word filtering model. The word filtering model may include filtering rules set according to a general thesaurus. In one embodiment, the computer device may obtain the text content included in the trademark of the first enterprise by: the computer device is text content included by a trademark of the first business using a text recognition algorithm. In one embodiment, the manner in which the computer device invokes the word filtering model to identify common words appearing in the text content may be: and the computer equipment calls a word filtering model to match each word in the text content with a general word bank so as to determine general words appearing in the text content, wherein the general word bank comprises a plurality of general words. Here, the filtering process may be understood as a deletion process.

In one embodiment, when the first business information includes a trademark of the first business, the computer device may extract the first key information for infringement determination from the first business information by: the computer device performs feature description on the image included in the trademark of the first enterprise by adopting an image feature description method to obtain key image information of the image included in the first trademark, and determines the key image information as first key information. In one embodiment, when the image characterization algorithm comprises a shape characterization algorithm, the computer device may characterize the image included in the first trademark using a shape characterization method, obtain shape features of the image included in the trademark of the first enterprise, and determine the shape features as the first key information.

In one embodiment, the first key information may include a keyword of a trademark of the first enterprise and/or key image information of the trademark of the first enterprise.

In one embodiment, when the first business information includes a business name of the first business, the computer device may extract the first key information for infringement determination from the first business information as follows: the computer equipment calls the word size extraction model to extract the enterprise word size included by the enterprise name of the first enterprise, and determines the enterprise word size included by the enterprise name of the first enterprise as first key information. The font size extraction model comprises extraction rules set according to the font size of the enterprise. In a practical application scenario, the process may be applied to a scenario where the first key information does not include the business word size of the first business. Since business names generally consist of four parts: the computer equipment can construct a corresponding regular expression according to the composition structure to serve as an extraction rule based on the enterprise font size, and the extraction rule based on the enterprise font size can be used for intercepting a part between an administrative division (such as Beijing city) and an industry (such as software) in the enterprise name, wherein the part is the enterprise font size. For example, the name of the first enterprise is Shanghai xx science and technology company, and the computer device can intercept xx from the xx science and technology company in Shanghai by calling the font size extraction model and determine xx as the first key information.

It should be noted that, the manner of extracting the second key information for infringement determination from the second enterprise information may refer to the manner of extracting the first key information for infringement determination from the first enterprise information, and details are not described herein.

Several ways of obtaining the comparison result by comparing the first key information and the second key information are explained below.

In one embodiment, when the first key information includes a keyword of a trademark of a first enterprise and the second key information includes a keyword of a trademark of a second enterprise, or when the first key information includes a business word size of the first enterprise and the second key information includes a business word size of the second enterprise, the method of the computer device comparing the first key information and the second key information may include at least one of: a comparison method based on character pronunciation and font, a comparison method based on character meaning and a comparison method based on character coincidence degree.

In one embodiment, when the method for comparing the first key information with the second key information includes a text meaning-based comparison method, the process of comparing the first key information with the second key information by the computer device to obtain the comparison result specifically includes: the computer equipment queries a plurality of pieces of information associated with the first key information based on the graph of the universal lexical synonymous features, queries whether the plurality of pieces of information associated with the first key information include second key information, determines that the first key information is similar to the second key information when the plurality of pieces of information associated with the first key information include the second key information, and obtains a first comparison result indicating that the first key information is similar to the second key information. A preliminary infringement determination result indicating that the first business information is suspected of infringing the second business information may then be obtained based on the first comparison result. The graph of the universal vocabulary synonymous characteristics can comprise a plurality of word nodes, and each word node is connected with the word node where the word synonymous with the word indicated by the word node is located.

In one embodiment, when the first key information includes key image information of a trademark of a first enterprise and the second key information includes key image information of a trademark of a second enterprise, the computer device compares the first key information and the second key information to obtain a comparison result specifically: the computer device may calculate a similarity between the key image information of the trademark of the first enterprise and the key image information of the trademark of the second enterprise, and when the similarity is greater than or equal to a preset value, determine that the first key information is similar to the second key information, and obtain a second comparison result indicating that the first key information is similar to the second key information. This method is suitable for the case where the trademark does not contain characters, or the trademark contains characters and images, and the shapes of the images included in the trademark are similar to each other although the characters are different from each other. A preliminary infringement determination may then be obtained indicating that the first business information is suspected of infringing the first business information.

In one embodiment, when the first key information includes a keyword of a trademark of a first enterprise and key image information of the trademark of the first enterprise, and the second key information includes a keyword of a trademark of a second enterprise and key image information of the trademark of the second enterprise, the computer device compares the first key information and the second key information to obtain a comparison result specifically by: the computer equipment queries a plurality of pieces of information related to the first key information based on the graph of the synonymy characteristic of the universal vocabulary, queries whether the plurality of pieces of information related to the first key information include second key information, and determines that the first key information is similar to the second key information when the plurality of pieces of information related to the first key information include the second key information, so as to obtain a first comparison result indicating that the first key information is similar to the second key information; and calculating the similarity between the key image information of the trademark of the first enterprise and the key image information of the trademark of the second enterprise, and determining that the first key information is similar to the second key information when the similarity is greater than or equal to a preset value to obtain a second comparison result indicating that the first key information is similar to the second key information. And then, obtaining a preliminary infringement judgment result indicating that the second enterprise information infringes the first enterprise information according to the first comparison result and the second comparison result.

And S104, searching article segments related to the first enterprise and the second enterprise.

In an embodiment of the application, the computer device may search for the article segments related to the first enterprise information and related to the second enterprise information by using a general search engine in data (such as news data) synchronized in real time on the internet. It should be noted that, in addition to performing a search based on data synchronized in real time on the internet, a search is performed based on data acquired in other manners (such as news data), which is not limited herein.

In one embodiment, to enable a quick search for article segments related to the first business information and related to the second business information, the computer device may search for article segments related to the first business information and related to the second business information using a general search engine based on the structured information of the synchronized data. In one embodiment, the computer device may extract structured information of the synchronized data using a general search engine and determine, from the synchronized data, article segments of which the structured information is first enterprise information and second enterprise information as the article segments related to the first enterprise and the second enterprise. The structured information may be structured information of the article title, the article author, and the like, which may carry enterprise information.

S105, carrying out sentiment classification on the article fragments and carrying out article property classification to obtain a classification result, wherein the classification result comprises a sentiment type and an article property type.

Wherein, the emotion category can be negative or positive, and the article property category can be promotional or non-promotional. The negative expression article segment may be an article segment which has a bad influence on the wiener enterprise image, such as rumor or libaran. The promotional properties represent the use of the article segments in commercial activities such as advertising. The nature of the promotion indicates a greater likelihood of infringement because the greater likelihood of false external assertions has a particular connection with the operator of a known good or service, with licensed uses, with associated businesses, etc. For example, the article author is related to the second enterprise information, the article title is related to the first enterprise information, and the article segment is an article segment which is not permitted by the right maintainer but is advertised by using the first enterprise information of the right maintainer. In this case, the computer device may determine that the article fragment is of a promotional nature by performing step S105.

In one embodiment, the computer device may perform sentiment classification and article property classification on the article segments by using the text classification model to obtain a classification result.

In one embodiment, the text classification model may be the first BERT (bidirectional Encoder expressions from transforms) model pre-trained for sentiment classification and article property classification. In one embodiment, since the initial BERT model only supports multi-level classification (only one label per sample), the initial BERT model may be modified to convert to the first BERT model that can support multi-label classification (multiple labels per sample). In particular, the classification function of the initial BERT model may be adjusted by the oftmax () function to one or more sigmoid () functions for multi-label classification. In addition, the loss function of the initial BERT model may be changed from-tf.reduce _ sum (one _ hot _ labels _ log _ probs, axis ═ 1) to tf.nn.sigmoid _ cross _ entry _ with _ locations (labels ═ labels, locations ═ locations). After the classification function and the loss function are adjusted for the initial BERT model, a first BERT model may be obtained.

In one embodiment, the text classification model may also include a pre-trained long-short term memory model for emotion classification and a pre-trained long-short term memory model for article property classification. The long-term and short-term memory model can be replaced by other neural network models capable of realizing the functions, and details are not repeated herein. Compared with the multi-label classification mode, the former multi-label classification mode can effectively avoid the problem of classification contradiction.

And S106, generating right-maintaining reference information according to the preliminary infringement judgment result and the classification result.

In the embodiment of the application, the computer device may generate right-maintaining reference information including a preliminary infringement judgment result and a classification result. The right reference information may be used as a clue for analyzing illicit competitive behavior related to intellectual property rights and/or may also be used as a clue for obtaining a final infringement judgment result. After that, the computer device can also obtain the judgment result of the illegal competition and/or obtain the final infringement judgment result according to the right-maintaining reference information by the illegal competition retrieval system. .

In one embodiment, the computer device may further perform named entity recognition on the text fragment to obtain a target named entity. The target named entity may include an article author, keywords related to the first business information, keywords related to the second business information, and so on. In one embodiment, the process of generating infringement determination information by the computer device according to the preliminary infringement determination result and the classification result may be: and the computer equipment generates right-maintaining reference information according to the preliminary infringement judgment result, the target named entity and the classification result. Here, the computer device may generate infringement determination information that includes the preliminary infringement determination result, the target named entity, and the classification result.

In one embodiment, the computer device may perform named entity recognition on the article fragment by using a named entity recognition model to obtain a target named entity. This named entity recognition model may be a second BERT model for named entity recognition that is pre-trained. The target named entity can be used to determine if infringement exists. For example, the computer device may perform named entity recognition on the text fragment a through a named entity recognition model to obtain a target named entity. Assume that the target named entity is: xxxx is a safe and healthy xxx. Although the target named entity contains the name of a brand of "safe", this "safe" in "safe health" is not meant by a company and thus does not constitute an infringement. Assume that the target named entity is: xxx, a target named entity, contains "safe" that is used as a company and thus has the possibility of infringement.

It can be seen that, in the embodiment shown in fig. 1, the computer device may acquire first enterprise information of a first enterprise and second enterprise information of a second enterprise, and extract first key information for infringement determination from the first enterprise information and extract second key information for infringement determination from the second enterprise information; the computer equipment compares the first key information with the second key information to obtain a comparison result, and determines a preliminary infringement judgment result according to the comparison result; the computer equipment searches article fragments related to a first enterprise and a second enterprise, and carries out sentiment classification on the article fragments and article property classification to obtain a classification result; the computer equipment generates right-maintaining reference information according to the preliminary infringement judgment result and the classification result; compared with the current process that the right is difficult to maintain if no one discovers or extracts the illegal competition and the infringement behaviors on a plurality of networks, the method and the device can discover the illegal competition and the infringement behaviors in time so as to maintain the right actively, and the right maintaining judgment information obtained by the method and the device can improve the data referential performance of the right maintaining process aiming at the illegal competition and the infringement behaviors.

Please refer to fig. 2, which is a flowchart illustrating an infringement processing method based on intelligent decision according to another embodiment of the present application. The method may be applied to a computer device. The computer device may be a server or an intelligent terminal. The server can be an independent server or a server cluster, and the intelligent terminal can be a desktop computer, a notebook computer and other intelligent terminals. Specifically, the method comprises the following steps:

s201, first enterprise information of a first enterprise and second enterprise information of a second enterprise are obtained.

S202, extracting first key information for infringement judgment from the first enterprise information, and extracting second key information for infringement judgment from the second enterprise information.

S203, comparing the first key information with the second key information to obtain a comparison result, and determining a preliminary infringement judgment result according to the comparison result.

And S204, searching article segments related to the first enterprise and the second enterprise.

S205, carrying out sentiment classification on the article fragments and carrying out article property classification to obtain classification results, wherein the classification results comprise sentiment categories and article property categories.

Steps S201 to S205 can refer to steps S101 to S105 in the embodiment of fig. 1, which are not described herein again.

S206, extracting a target triple from the article fragment, wherein the target triple is composed of an entity of a first enterprise, an entity of a second enterprise and a first incidence relation between the first enterprise and the second enterprise.

In this embodiment, the computer device may extract a target triple in the article fragment by using a triple extraction model, where the target triple is formed by the first enterprise, the second enterprise, and the first association relationship between the first enterprise and the second enterprise. This triplet extraction model may be a third BERT model for triplet extraction that is pre-trained.

In one embodiment, the way that the computer device extracts the target triples in the article fragment by using the triple extraction model may be: the computer equipment extracts the semantic features of the article fragments through the triple extraction model, and extracts the features of the entity of the first enterprise and the entity of the second enterprise of the article fragments; the computer device determines a first incidence relationship between the first business and the second business according to the extracted semantic features, the features of the entity of the first business and the features of the entity of the second business. And further obtaining a target triple formed by the entity of the first enterprise, the entity of the second enterprise and the first incidence relation.

And S207, obtaining a authenticity verification result of the first association relation according to the enterprise knowledge graph and the target triple.

An enterprise knowledge graph may include, among other things, a plurality of enterprise nodes and associations between the plurality of enterprise nodes (which may be, for example, some type of partnership). Generally, the correlation is false, with a greater likelihood of infringement. In one embodiment, each enterprise node in the enterprise knowledge graph may also have connected thereto an event node or the like.

In this embodiment of the application, the manner in which the computer device obtains the authenticity verification result of the first association relationship based on the enterprise knowledge graph and the target triple may be: the computer equipment queries at least one enterprise related to the first enterprise based on the enterprise knowledge graph, judges whether the at least one enterprise comprises a second enterprise, extracts a second association relation between the first enterprise and the second enterprise from the enterprise knowledge graph when the at least one enterprise comprises the second enterprise, verifies whether the first enterprise relation and the second enterprise relation are consistent, if so, obtains a true-false verification result indicating that the first association relation is true, and if not, determines that the first association relation is false. Generally, the relationship is false, with a greater likelihood of infringement. According to the embodiment of the application, the authenticity degree of the relation can be rapidly positioned through a searching mode based on the enterprise knowledge graph.

And S208, generating right-maintaining reference information according to the preliminary infringement judgment result, the classification result and the authenticity verification result of the first association relation.

In this embodiment, the computer device may generate a preliminary infringement determination result, a classification result, and right-maintaining reference information for the first association. The right reference information may be used as a clue for analyzing illicit competitive behavior related to intellectual property rights and/or may also be used as a clue for obtaining a final infringement judgment result. After that, the computer device can also obtain the judgment result of the illegal competition and/or obtain the final infringement judgment result according to the right-maintaining reference information by the illegal competition retrieval system.

In one embodiment, the computer device may further extract target event information from the article segment, extract an event information set from the enterprise knowledge graph, where the event information set includes first event information associated with a first enterprise and second event information associated with a second enterprise, further query whether the event information set includes the target event information, and obtain a result of verifying the authenticity of the target event information according to the query result. The first event information refers to event information associated with a first enterprise, and the second event information refers to event information associated with a second enterprise. In one embodiment, if the query result indicates that the set of event information includes the target event information, a true-false verification result indicating that the target event information is true is obtained, and if the query result indicates that the set of event information does not include the target event information, a true-false verification result indicating that the target event information is false is obtained. Generally, an event is false, with a greater likelihood of infringement. According to the method and the device, the authenticity degree of the event can be rapidly positioned through a searching mode based on the enterprise knowledge graph. Correspondingly, the process of generating the right-maintaining reference information by the computer device according to the preliminary infringement judgment result and the classification result may be: and the computer equipment generates right-maintaining reference information according to the preliminary infringement judgment result, the classification result and the authenticity verification result of the target event information. Specifically, the computer device may generate infringement determination information including a preliminary infringement determination result, a classification result, and a result of authenticity verification of the target event information.

In one embodiment, the infringement determination result generated by the computer device may include a preliminary infringement determination result, a target named entity, a classification result, and a result of verifying authenticity of the first association relation, or the infringement determination result may further include a preliminary infringement determination result, a target named entity, a classification result, a result of verifying authenticity of the first association relation, and a result of verifying authenticity of the target event information.

In one embodiment, the manner in which the computer device obtains the first enterprise information of the first enterprise may further be: the method comprises the steps that a computer device obtains an enterprise information set of a first enterprise, wherein the enterprise information set comprises a plurality of enterprise information; the computer device screens out first business information from a plurality of business information. In one embodiment, the computer device may filter out the first business information from the plurality of business information by: the computer device screens out at least one enterprise information in a language of Chinese from the plurality of enterprise information, and screens out enterprise information belonging to a target category from the at least one enterprise information as first enterprise information. For example, the plurality of business information includes a first number of trademarks, and the computer device may screen at least one trademark in chinese from the first number of trademarks and then screen trademarks belonging to the target category from the at least one trademark as the first trademark. In one embodiment, the target category may be any one or more of: trade mark, service mark, collective mark, and certification mark. The target category may be obtained by adopting a classification manner according to other classification manners, in addition to the classification manner according to the trademark use object, which is not described herein again. Further, when the object category is plural and each object category is different, the trademark belonging to the object category refers to a trademark belonging to each of the plural object categories. In one embodiment, the computer device may output a search interface provided by a general search engine through the terminal device, such that the terminal device submits the set of enterprise information based on the search interface.

In one embodiment, the manner in which the computer device obtains the second enterprise information for the second enterprise may be as follows: the method comprises the steps that computer equipment filters an enterprise information base according to target enterprise information of a first enterprise and a filtering rule set based on the enterprise information to obtain a first enterprise information set, wherein the enterprise information base comprises a plurality of enterprise information, and the first enterprise information set comprises at least one piece of enterprise information with the first character being the same as that of the target enterprise information; and the computer equipment determines the operation range of the first enterprise according to the target enterprise information, and filters the first enterprise information set according to the operation range of the first enterprise and a filtering rule set based on the operation range to obtain a second enterprise information set, wherein the second enterprise information set comprises enterprise information of a second enterprise, the operation range of which is consistent with that of the first enterprise. The target enterprise information may include a word size of the first enterprise or an abbreviation of the first enterprise. In an embodiment, the computer device may further perform filtering of the enterprise information in combination with the filtering rule based on the enterprise white list, which is not described herein in detail. Wherein the enterprise filtering rules may be executed by the server through a general search engine. According to the embodiment of the application, the enterprise filtering rule is added on the basis of the search technology of the universal search engine, so that the magnitude order of search can be effectively reduced, and particularly, the enterprise information search of tens of millions of levels is realized. The prior art generally searches through a general search engine. For example, for the universal search engine, the search principle is mainly based on the inverted index technology, and the participle index is used for the participle query of the Chinese to accelerate the search speed. However, the searching method cannot meet the query requirements of trademarks and enterprise word sizes with similar characters, sounds, characters and shapes, similar meanings and overlapped partial characters. Moreover, most trademarks and enterprise names belong to brand-new creative vocabularies, and can not be correctly segmented. Therefore, the search technology based on the universal search engine cannot search the infringers, and the subsequent text comparison is needless to say. Therefore, in a practical application scenario, the embodiment of the present application may configure the enterprise filtering rules into a server or a database by the relevant personnel based on a search page provided by a general search engine or based on a front-end page provided by a computer device. Wherein the enterprise filtering rules are at least one of the following: the filtering rules are based on enterprise abbreviation or enterprise font size, the filtering rules are based on enterprise operation range, and the filtering rules are based on enterprise white list. Based on the filtering rule of the enterprise abbreviation or the enterprise font size, the registration information of the enterprise with the first character of the enterprise abbreviation or the first character of the enterprise font size different from the right-maintaining party can be filtered. Based on the filtering rule of the enterprise operation range, the registration information of the enterprise with the operation range different from the right-maintaining party can be filtered. Registration information for the enterprises in the enterprise whitelist may be filtered based on the filtering rules of the enterprise whitelist.

In an embodiment, the present application relates to a block chain technology, for example, right-to-maintain reference information may be written in a block chain, and since data of the block chain is not easily tampered, the information may be well verified. In one embodiment, the first enterprise information and/or the second enterprise information may be read from the blockchain or may be broadcast to the computer device through a blockchain network. In one embodiment, the first enterprise information may also be obtained by the computer device from a node device of the first enterprise (e.g., may be a node device in a blockchain network). The second enterprise information may also be obtained by the computer device from a node device of the second enterprise (e.g., may be a node device in a blockchain network).

It can be seen that, in the embodiment shown in fig. 2, the computer device may further verify a first association relationship between the first enterprise and the second enterprise, and obtain an authenticity determination result for the first association relationship; and then generating the right-maintaining reference information according to the preliminary infringement judgment result, the classification result and the authenticity judgment result of the first association relation, enriching the dimensionality of the right-maintaining reference information and facilitating the subsequent analysis of illegal competition and infringement behaviors.

Please refer to fig. 3, which is a schematic structural diagram of an infringement processing apparatus based on intelligent decision according to an embodiment of the present application. The infringement processing apparatus may be applied to the aforementioned computer device. Specifically, the infringement processing apparatus may include:

the obtaining module 301 is configured to obtain first enterprise information of a first enterprise and second enterprise information of a second enterprise.

An extracting module 302, configured to extract first key information for infringement determination from the first enterprise information, and extract second key information for infringement determination from the second enterprise information.

A comparison module 303, configured to compare the first key information and the second key information to obtain a comparison result, and determine a preliminary infringement determination result according to the comparison result.

A search module 304 is configured to search for article snippets that are related to the first business and related to the second business.

The classification module 305 is configured to perform emotion classification on the article segments and perform article property classification to obtain classification results, where the classification results include an emotion category and an article property category.

And an information generating module 306, configured to generate right-maintaining reference information according to the preliminary infringement determination result and the classification result.

In an optional implementation manner, the first enterprise information includes a trademark of the first enterprise, and the extracting module 302 is specifically configured to:

In an optional implementation manner, the first enterprise information includes an enterprise name of the first enterprise, and the extracting module 302 is specifically configured to:

In an optional implementation manner, the obtaining module 301 is specifically configured to:

In one embodiment, the infringement processing apparatus further includes an entity identification module 307.

In an embodiment, the entity identifying module 307 is configured to perform named entity identification on the text fragment to obtain a target named entity.

In one embodiment, the information generating module 306 is specifically configured to:

In one embodiment, the infringement handling device further includes a relationship verification module 308.

In one embodiment, the relationship verification module 308 is specifically configured to:

and acquiring a true and false verification result of the first association relation according to the enterprise knowledge graph and the target triple.

In one embodiment, the infringement processing device further includes an event verification module 309.

In one embodiment, the event verification module 309 is specifically configured to:

extracting target event information from the article fragment;

and inquiring whether the event information set comprises the target event information or not, and obtaining a true-false verification result of the target event information according to an inquiry result.

It can be seen that, in the embodiment shown in fig. 3, the computer device may acquire first enterprise information of a first enterprise and second enterprise information of a second enterprise, and extract first key information for infringement determination from the first enterprise information and extract second key information for infringement determination from the second enterprise information; the computer equipment compares the first key information with the second key information to obtain a comparison result, and determines a preliminary infringement judgment result according to the comparison result; the computer equipment searches article fragments related to a first enterprise and a second enterprise, and carries out sentiment classification on the article fragments and article property classification to obtain a classification result; the computer equipment generates the right maintaining reference information according to the preliminary infringement judgment result and the classification result, so that the illegal competition and the infringement behavior can be found in time to maintain the right actively, and the right maintaining judgment information obtained by the application can improve the data referential performance of the right maintaining process aiming at the illegal competition and the infringement behavior.

Please refer to fig. 4, which is a schematic structural diagram of a computer device according to an embodiment of the present application. The computer device described in this embodiment may include: one or more processors 1000 and memory 2000. The processor 1000 and the memory 2000 may be connected by a bus.

The Processor 1000 may be a Central Processing Unit (CPU), and may be other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 2000 may be a high-speed RAM memory or a non-volatile memory (e.g., a disk memory). Wherein the memory 2000 is used for storing a computer program comprising program instructions, the processor 1000 is configured for invoking the program instructions for performing the steps of:

In one embodiment, the first enterprise information includes a trademark of a first enterprise, and when first key information for infringement determination is extracted from the first enterprise information, the processor 1000 is configured to invoke the program instructions to perform the following steps:

acquiring text content included by the trademark of the first enterprise;

In one embodiment, the first enterprise information includes an enterprise name of the first enterprise, and when first key information for infringement determination is extracted from the first enterprise information, the processor 1000 is configured to invoke the program instructions to perform the following steps:

In one embodiment, in obtaining second enterprise information for a second enterprise, the processor 1000 is configured to invoke the program instructions to perform the steps of:

In one embodiment, the processor 1000 is configured to invoke the program instructions and further perform the steps of:

and carrying out named entity recognition on the text fragment to obtain a target named entity.

In one embodiment, when generating the right-maintaining reference information according to the preliminary infringement determination result and the classification result, the processor 1000 is configured to call the program instruction to perform the following steps:

extracting target event information from the article fragment;

In a specific implementation, the processor 1000 described in this embodiment of the present application may execute the implementation described in the embodiment of fig. 1 and the embodiment of fig. 2, and may also execute the implementation described in this embodiment of the present application, which is not described herein again.

The functional modules in the embodiments of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a form of sampling hardware, and can also be realized in a form of sampling software functional modules.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The computer readable storage medium may be volatile or nonvolatile. For example, the computer storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like. The computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the blockchain node, and the like.

The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism and an encryption algorithm. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.

While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. An infringement processing method based on intelligent decision is characterized by comprising the following steps:

2. The method of claim 1, wherein the first business information includes a trademark of a first business, and wherein extracting first key information for infringement determination from the first business information includes:

acquiring text content included by the trademark of the first enterprise;

3. The method of claim 1, wherein the first business information comprises a business name of the first business, and wherein extracting first key information for infringement determination from the first business information comprises:

4. The method of claim 1, wherein obtaining second business information for a second business comprises:

5. The method of claim 1, further comprising:

6. The method of claim 1, further comprising:

7. The method of claim 1, further comprising:

extracting target event information from the article fragment;

8. An infringement processing device based on intelligent decision-making, comprising:

9. A computer device comprising a processor and a memory, the processor and the memory being interconnected, wherein the memory is configured to store a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which is executed by a processor to implement the method according to any one of claims 1-7.