A patent recommendation method and system for customers

Abstract

This invention provides a patent recommendation method and system for customers, relating to the field of computer technology. The method includes: acquiring customer demand information and / or customer information; determining filtering rules based on the demand information and / or customer information; filtering at least one patent text from a database based on the filtering rules; and constructing and outputting a recommendation interface based on the filtered patent text. This invention's customer-oriented patent recommendation method enables patent recommendations to be made according to customer requirements.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a patent recommendation method and system for customers. Background Technology

[0002] Intellectual property, as an intangible asset of enterprises, is receiving increasing attention from businesses. Intellectual property includes copyright, patent rights, trademark rights, discovery rights, invention rights, and other rights to scientific and technological achievements. Works, trademarks, trade secrets, inventions, utility models, and designs are common intellectual property objects; among them, patents (inventions, utility models, and designs) are a highly representative form of intellectual property. Given the vast amount of existing patent data, and the fact that most clients lack in-depth expertise in patents, it is essential to recommend patents to clients to ensure the effective application of patented technologies. Summary of the Invention

[0003] One of the objectives of this invention is to provide a patent recommendation method for customers, enabling patent recommendations to be made based on customer requirements.

[0004] This invention provides a patent recommendation method for customers, comprising:

[0005] Obtain customer needs information and / or customer information;

[0006] Based on the aforementioned demand information and / or customer information, the screening rules are determined;

[0007] Based on the filtering rules, at least one patent text is selected from the database;

[0008] Based on the selected patent texts, a recommendation interface is constructed and output.

[0009] Preferably, obtaining customer demand information and / or customer information includes:

[0010] Output the preset requirement information retrieval page;

[0011] Receive input information from customers on the demand information acquisition page for each demand item;

[0012] And / or,

[0013] Output the preset customer entity name input page;

[0014] Receive the customer entity name entered by the user on the customer entity name input page;

[0015] Based on the customer's name, the customer information is obtained from a pre-set third-party platform.

[0016] Preferably, determining the screening rules based on the demand information and / or the customer information includes:

[0017] Analyze the requirement information to determine the input information for each requirement item;

[0018] Obtain the first quantization template corresponding to each of the aforementioned requirement items;

[0019] Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined;

[0020] Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values;

[0021] Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed;

[0022] Retrieve the preset filtering rule library;

[0023] The filtering rules are determined based on the extracted parameter set and the filtering rule base.

[0024] Preferred, customer-oriented patent recommendation methods also include:

[0025] After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations.

[0026] When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface.

[0027] Receive recommendation and analysis information from various experts for each of the aforementioned patent texts;

[0028] When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information.

[0029] The recommendation interface is updated based on the expert recommendation score.

[0030] Preferably, determining the expert recommendation level for each patent text based on the recommendation analysis information includes:

[0031] Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value;

[0032] Determine the relationship between each expert and the patent text;

[0033] Based on the aforementioned association relationships and the preset association relationship and weight correspondence table, the weight corresponding to each expert is determined;

[0034] The expert recommendation degree of the patent text is determined based on the weight of each expert and its corresponding second quantitative value.

[0035] This invention also provides a customer-oriented patent recommendation system, comprising:

[0036] The acquisition module is used to acquire customer demand information and / or customer information;

[0037] The rule determination module is used to determine the screening rules based on the demand information and / or customer information;

[0038] A filtering module is used to filter at least one patent text from the database based on the filtering rules;

[0039] A construction module is used to build and output a recommendation interface based on the selected patent texts.

[0040] Preferably, the acquisition module acquires customer demand information and / or customer information, and performs the following operations:

[0041] Output the preset requirement information retrieval page;

[0042] Receive input information from customers on the demand information acquisition page for each demand item;

[0043] And / or,

[0044] Output the preset customer entity name input page;

[0045] Receive the customer entity name entered by the user on the customer entity name input page;

[0046] Based on the customer's name, the customer information is obtained from a pre-set third-party platform.

[0047] Preferably, the rule determination module determines the filtering rules based on the demand information and / or the customer information, and performs the following operations:

[0048] Analyze the requirement information to determine the input information for each requirement item;

[0049] Obtain the first quantization template corresponding to each of the aforementioned requirement items;

[0050] Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined;

[0051] Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values;

[0052] Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed;

[0053] Retrieve the preset filtering rule library;

[0054] The filtering rules are determined based on the extracted parameter set and the filtering rule base.

[0055] Preferably, a customer-oriented patent recommendation system also includes: an expert recommendation and analysis module;

[0056] The expert recommendation analysis module performs the following operations:

[0057] After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations.

[0058] When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface.

[0059] Receive recommendation and analysis information from various experts for each of the aforementioned patent texts;

[0060] When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information.

[0061] The recommendation interface is updated based on the expert recommendation score.

[0062] Preferably, the expert recommendation analysis module determines the expert recommendation degree of each patent text based on the recommendation analysis information and performs the following operations:

[0063] Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value;

[0064] Determine the relationship between each expert and the patent text;

[0065] Based on the aforementioned association relationships and the preset association relationship and weight correspondence table, the weight corresponding to each expert is determined;

[0066] The expert recommendation degree of the patent text is determined based on the weight of each expert and its corresponding second quantitative value.

[0067] The present invention has the following beneficial effects:

[0068] 1. Make adaptive patent recommendations based on the client's needs and / or client information;

[0069] Second, by publicizing the recommendation interface and the information used to build the recommendation interface, and by incorporating expert opinions, we provide guidance based on expert recommendations, making it easier for clients to obtain the best patent texts, so that they can make further arrangements to obtain licenses or transfers.

[0070] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0071] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0072] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0073] Figure 1 This is a schematic diagram of a patent recommendation method for customers according to an embodiment of the present invention;

[0074] Figure 2 This is a schematic diagram of a patent recommendation method for customers according to an embodiment of the present invention. Detailed Implementation

[0075] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0076] This invention provides a patent recommendation method for customers, such as... Figure 1 As shown, it includes:

[0077] Step S1: Obtain customer demand information and / or customer information; demand information includes: patent field, patent keywords, patent transfer price, etc.; customer information includes: customer name, business scope, products sold, etc.

[0078] Step S2: Determine the screening rules based on demand information and / or customer information; realize the determination of appropriate screening rules based on demand information and / or customer information in order to select suitable patents from the database;

[0079] Step S3: Based on the filtering rules, at least one patent text is selected from the database; for example, the filtering rule is: keyword filtering, the keyword is XX; the patent texts in the database containing XX are selected; another example is: the filtering rule is: patent text matching, the matching patent text is A, that is, the patent texts in the database that match patent text A are selected; in addition, the filtering rules also include: classification number filtering, patent name filtering, etc.; the database includes the total database and the sub-databases containing transferable or licenseable patents;

[0080] Step S4: Based on the selected patent texts, construct and output a recommendation interface. Clients can intuitively view the recommended patents through this interface. The patents on the recommendation interface are sorted according to their fit with the selection rules; when the fit is the same, they are sorted by application date. The recommendation interface uses a preset recommendation template and is generated by collecting corresponding information from the patent texts. Each recommended patent text has its corresponding area on the recommendation interface, displaying information including: the first attached drawing, patent name, bibliographic information, and abstract.

[0081] In one embodiment, obtaining customer demand information and / or customer information includes:

[0082] Output the preset requirement information retrieval page;

[0083] Receive input information from customers on the demand information acquisition page for each demand item; demand items include: the price range of the transfer to be received, the field involved, the application date range, the patent status, etc.

[0084] And / or,

[0085] Output the preset customer entity name input page;

[0086] Receive the customer entity name entered by the user on the customer entity name input page;

[0087] Based on the customer's entity name, the customer information is obtained from a pre-defined third-party platform. This third-party platform includes patent database platforms, enterprise information query platforms, etc.; for example, logging into the website corresponding to the customer's entity name through a third-party platform to obtain data such as the products and business scope on the website.

[0088] In one embodiment, determining the screening rules based on the demand information and / or the customer information includes:

[0089] Analyze the requirement information to determine the input information for each requirement item;

[0090] Obtain the first quantization template corresponding to each of the aforementioned requirement items;

[0091] Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined; the input information of the requirement items is quantized into the first quantization value to facilitate the determination of subsequent filtering rules.

[0092] Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values;

[0093] Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed; each of the quantized values ​​and / or multiple of the feature values ​​is filled into the blank template of the extraction parameter set to realize the construction of the extraction parameter set;

[0094] Retrieve the preset filtering rule library;

[0095] Based on the extracted parameter set and the filtering rule base, the filtering rules are determined. The extracted parameter set is matched with the identifier parameter set corresponding to each filtering rule in the filtering rule base, and the filtering rules corresponding to the matching identifier parameter sets are extracted. The matching can be achieved by calculating the similarity between the extracted parameters and the identifier parameter sets; that is, when the similarity is the highest in the filtering rule base, the two are matched.

[0096] In one embodiment, the customer-oriented patent recommendation method further includes:

[0097] After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations.

[0098] When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface.

[0099] Receive recommendation and analysis information from various experts for each of the aforementioned patent texts;

[0100] When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information.

[0101] The recommendation interface is updated based on the expert recommendation score.

[0102] The step of determining the expert recommendation level for each patent text based on the recommendation analysis information includes:

[0103] Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value;

[0104] Determine the relationship between each expert and the patent text; the relationship is considered relevant if the agency employed by the expert is the same as the agency employed by the patent text; otherwise, it is considered irrelevant.

[0105] Based on the aforementioned correlation and the preset correlation and weight correspondence table, the weight corresponding to each expert is determined; when they are related, the weight is zero; when they are not related, the weight is 1.

[0106] The expert recommendation score of the patent text is determined based on the weights corresponding to each expert and their corresponding second quantification values. The average of the products of the second quantification values ​​and the weights is taken as the expert recommendation score.

[0107] The working principle and beneficial effects of the above technical solution are as follows:

[0108] The system packages the demand information and / or customer information, the selected patent texts, and the recommendation interface to generate an analysis request, which is then displayed in the expert joint analysis area. This area is a system-defined region for displaying patent recommendation schemes awaiting expert analysis. Users can determine the termination of data collection by setting a display period or a threshold for the number of expert analyses. After termination, the system comprehensively analyzes the expert recommendations, providing further guidance to customers on patent acquisition and other operations. Furthermore, when updating the recommendation interface, the system extracts the expert information that provided the recommendation analysis. If a well-known expert is pre-stored in the system, the expert's watermark is extracted and added to the display area of ​​the corresponding patent on the recommendation interface, and a "Known" label is added to the upper right corner of the display area using a preset annotation method. The outer frame of the display area is highlighted with a bright border. Additionally, based on the expert recommendation score, the recommendation interface is updated, including displaying the expert recommendation score for each patent text in the corresponding recommendation area.

[0109] This invention also provides a customer-oriented patent recommendation system, such as... Figure 2 As shown, it includes:

[0110] Module 1 is used to acquire customer demand information and / or customer information;

[0111] Rule determination module 2 is used to determine screening rules based on the demand information and / or customer information;

[0112] Filtering module 3 is used to filter at least one patent text from the database based on the filtering rules;

[0113] Module 4 is used to build and output a recommendation interface based on the selected patent texts.

[0114] Preferably, the acquisition module 1 acquires customer demand information and / or customer information, and performs the following operations:

[0115] Output the preset requirement information retrieval page;

[0116] Receive input information from customers on the demand information acquisition page for each demand item;

[0117] And / or,

[0118] Output the preset customer entity name input page;

[0119] Receive the customer entity name entered by the user on the customer entity name input page;

[0120] Based on the customer's name, the customer information is obtained from a pre-set third-party platform.

[0121] Preferably, the rule determination module 2 determines the screening rules based on the demand information and / or the customer information, and performs the following operations:

[0122] Analyze the requirement information to determine the input information for each requirement item;

[0123] Obtain the first quantization template corresponding to each of the aforementioned requirement items;

[0124] Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined;

[0125] Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values;

[0126] Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed;

[0127] Retrieve the preset filtering rule library;

[0128] The filtering rules are determined based on the extracted parameter set and the filtering rule base.

[0129] Preferably, a customer-oriented patent recommendation system also includes: an expert recommendation and analysis module;

[0130] The expert recommendation analysis module performs the following operations:

[0131] After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations.

[0132] When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface.

[0133] Receive recommendation and analysis information from various experts for each of the aforementioned patent texts;

[0134] When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information.

[0135] The recommendation interface is updated based on the expert recommendation score.

[0136] Preferably, the expert recommendation analysis module determines the expert recommendation degree of each patent text based on the recommendation analysis information and performs the following operations:

[0137] Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value;

[0138] Determine the relationship between each expert and the patent text;

[0139] Based on the aforementioned association relationships and the preset association relationship and weight correspondence table, the weight corresponding to each expert is determined;

[0140] The expert recommendation degree of the patent text is determined based on the weight of each expert and its corresponding second quantitative value.

[0141] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A patent recommendation method for customers, characterized in that, include: Obtain customer needs information and / or customer information; Based on the aforementioned demand information and / or customer information, the screening rules are determined; Based on the filtering rules, at least one patent text is selected from the database; Based on the selected patent texts, a recommendation interface is constructed and output; Specifically, it also includes: After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations. When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface. Receive recommendation and analysis information from various experts for each of the aforementioned patent texts; When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information. The recommendation interface is updated based on the expert recommendation score. The step of determining the expert recommendation level for each patent text based on the recommendation analysis information includes: Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value; Determine the relationship between each expert and the patent text; the relationship is considered relevant if the agency employed by the expert is the same as the agency employed by the patent text; otherwise, it is considered irrelevant. Based on the aforementioned correlation and the preset correlation and weight correspondence table, the weight corresponding to each expert is determined; when they are related, the weight is zero; when they are not related, the weight is 1. Based on the weights corresponding to each expert and their corresponding second quantification values, the expert recommendation degree of the patent text is determined, and the average of the product of the second quantification value and the weight is used as the expert recommendation degree.

2. The customer-oriented patent recommendation method as described in claim 1, characterized in that, The acquisition of customer demand information and / or customer information includes: Output the preset requirement information retrieval page; Receive input information from customers on the demand information acquisition page for each demand item; And / or, Output the preset customer entity name input page; Receive the customer entity name entered by the user on the customer entity name input page; Based on the customer's name, the customer information is obtained from a pre-set third-party platform.

3. The customer-oriented patent recommendation method as described in claim 1, characterized in that, The step of determining the screening rules based on the demand information and / or the customer information includes: Analyze the requirement information to determine the input information for each requirement item; Obtain the first quantization template corresponding to each of the aforementioned requirement items; Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined; Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values; Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed; Retrieve the preset filtering rule library; The filtering rules are determined based on the extracted parameter set and the filtering rule base.

4. A customer-oriented patent recommendation system, characterized in that, include: The acquisition module is used to acquire customer demand information and / or customer information; The rule determination module is used to determine the screening rules based on the demand information and / or customer information; A filtering module is used to filter at least one patent text from the database based on the filtering rules; A construction module is used to build and output a recommendation interface based on the selected patent texts; Specifically, it also includes: an expert recommendation analysis module; The expert recommendation analysis module performs the following operations: After the recommendation interface is output, a preset question is output asking the customer whether they want to receive expert analysis and recommendations. When a customer confirms their response to the inquiry, an analysis request is generated and published in the expert joint analysis area based on the demand information and / or customer information, the selected patent text, and the recommendation interface. Receive recommendation and analysis information from various experts for each of the aforementioned patent texts; When the public notice period expires or a predetermined number of the recommended analysis information is received, the expert recommendation degree of each patent text is determined based on the recommended analysis information. The recommendation interface is updated based on the expert recommendation score. The step of determining the expert recommendation level for each patent text based on the recommendation analysis information includes: Based on a preset second quantization template, the recommendation analysis information is quantified to obtain a second quantization value; Determine the relationship between each expert and the patent text; the relationship is considered relevant if the agency employed by the expert is the same as the agency employed by the patent text; otherwise, it is considered irrelevant. Based on the aforementioned correlation and the preset correlation and weight correspondence table, the weight corresponding to each expert is determined; when they are related, the weight is zero; when they are not related, the weight is 1. Based on the weights corresponding to each expert and their corresponding second quantification values, the expert recommendation degree of the patent text is determined, and the average of the product of the second quantification value and the weight is used as the expert recommendation degree.

5. The customer-oriented patent recommendation system as described in claim 4, characterized in that, The acquisition module acquires customer demand information and / or customer information, and performs the following operations: Output the preset requirement information retrieval page; Receive input information from customers on the demand information acquisition page for each demand item; And / or, Output the preset customer entity name input page; Receive the customer entity name entered by the user on the customer entity name input page; Based on the customer's name, the customer information is obtained from a pre-set third-party platform.

6. The customer-oriented patent recommendation system as described in claim 4, characterized in that, The rule determination module determines the screening rules based on the demand information and / or the customer information, and performs the following operations: Analyze the requirement information to determine the input information for each requirement item; Obtain the first quantization template corresponding to each of the aforementioned requirement items; Based on the input information corresponding to each of the aforementioned requirement items and the first quantization template, each first quantization value is determined; Based on a preset feature extraction template, feature extraction is performed on the customer information to obtain multiple feature values; Based on each of the quantized values ​​and / or multiple of the feature values, an extraction parameter set is constructed; Retrieve the preset filtering rule library; The filtering rules are determined based on the extracted parameter set and the filtering rule base.

Images