Patent search method and device, and storage medium

Abstract

The application provides a patent retrieval method and device and a storage medium. The most suitable document set satisfying a retrieval parameter combination condition can be quickly found by browsing a label of the retrieval parameter combination, and then corresponding patent documents are retrieved, so that the retrieval time is greatly reduced and the retrieval efficiency is improved. The method comprises the following steps: providing a retrieval interface; determining a retrieval condition in the retrieval interface according to a condition determination operation instruction of a user; determining a retrieval parameter corresponding to the retrieval condition, wherein the retrieval parameter comprises at least one of a keyword, an IPC and a CPC; determining P retrieval parameter combinations corresponding to the retrieval parameter, wherein P is an integer greater than or equal to 1; and performing retrieval on a target data set based on the P retrieval parameter combinations to obtain a patent document set hit by each of the P retrieval parameter combinations, wherein the target data set corresponds to the retrieval condition.

Description

Technical Field

[0001] This application belongs to the field of search, and specifically relates to a patent search method, apparatus and storage medium. Background Technology

[0002] Traditional Boolean retrieval relies on Boolean logic, specifically keyword-based searches, to locate the documents being searched. Due to the binary nature of Boolean logic, users cannot use too many keywords to avoid missing results. However, to discover suitable and relevant documents, users use semantic ranking to prioritize the most relevant documents.

[0003] Currently, users can only use a limited number of keywords to locate corresponding search documents, and semantic ranking can only provide a scalar (numerical) relevance score of 0-100%. Besides indicating that these documents meet the Boolean search criteria, it cannot provide further information about the documents. Therefore, users have a significant reading task, wasting a lot of time trying to find more patent documents that match their target keywords (technical elements). Summary of the Invention

[0004] This application provides a patent search method, apparatus, and storage medium that allows users to browse parameter tags for search parameter combinations, find the most suitable combination of search parameters, and then retrieve the corresponding patent documents, greatly reducing search time.

[0005] The first aspect of this application provides a patent search method, including:

[0006] Provide a search interface;

[0007] Based on the user's criteria, the operation instructions are determined on the search interface, specifying the search conditions.

[0008] Determine the search parameters corresponding to the search criteria, wherein the search parameters include at least one of keywords, IPC, and CPC;

[0009] Determine P combinations of search parameters corresponding to the search parameters, where P is an integer greater than or equal to 1;

[0010] The target dataset is searched based on the P search parameter combinations to obtain the set of patent documents matched by each of the P search parameter combinations, and the target dataset corresponds to the search conditions.

[0011] A second aspect of this application provides a patent search device, comprising:

[0012] The interface providing unit is used to provide a search interface;

[0013] The first determining unit is used to determine the operation instructions based on the user's conditions and to determine the search conditions on the search interface.

[0014] The second determining unit is used to determine the search parameters corresponding to the search conditions, wherein the search parameters include at least one of keywords, IPC, and CPC;

[0015] A combination unit is used to determine P combinations of search parameters corresponding to the search parameters, wherein P is an integer greater than or equal to 1;

[0016] The retrieval unit is used to retrieve the target dataset based on the P retrieval parameter combinations to obtain the set of patent documents matched by each of the P retrieval parameter combinations, wherein the target dataset corresponds to the retrieval conditions.

[0017] A third aspect of this application provides a computer device comprising at least one connected processor, memory, and transceiver, wherein the memory is used to store program code, and the processor is used to invoke the program code in the memory to execute the steps of the patent search method described in the first aspect.

[0018] The fourth aspect of this application provides a computer storage medium including instructions that, when executed on a computer, cause the computer to perform the steps of the patent search method described in the first aspect.

[0019] Compared to related technologies, the embodiments provided in this application involve a patent search device determining search conditions on the search interface based on user-defined operation instructions; determining search parameters corresponding to the search conditions, including keywords, IPC, and CPC; arbitrarily combining keywords, IPC, and CPC in the search parameters to obtain P search parameter combinations; and searching the target dataset based on the P search parameter combinations to obtain the patent documents matched by each of the P search parameter combinations. Therefore, users do not need to read every patent document retrieved; they can find the most suitable search parameter combination by browsing the search parameter combination tags, thereby retrieving the corresponding patent documents, greatly reducing search time. Attached Figure Description

[0020] Figure 1 A schematic flowchart illustrating the patent search method provided in this application embodiment;

[0021] Figure 2A A schematic diagram of the search interface provided in an embodiment of this application;

[0022] Figure 2B Another schematic diagram of the search interface provided in the embodiments of this application;

[0023] Figure 2C Another schematic diagram of the search interface provided in the embodiments of this application;

[0024] Figure 2D Another schematic diagram of the search interface provided in the embodiments of this application;

[0025] Figure 3A A schematic diagram of the interface provided for the embodiment of this application;

[0026] Figure 3B Another schematic diagram of the display interface provided in the embodiments of this application;

[0027] Figure 3C Another schematic diagram of the display interface provided in the embodiments of this application;

[0028] Figure 3D Another schematic diagram of the display interface provided in the embodiments of this application;

[0029] Figure 4A Another schematic diagram of the display interface provided in the embodiments of this application;

[0030] Figure 4B Another schematic diagram of the display interface provided in the embodiments of this application;

[0031] Figure 4C Another schematic diagram of the display interface provided in the embodiments of this application;

[0032] Figure 5A Another schematic diagram of the display interface provided in the embodiments of this application;

[0033] Figure 5B Another schematic diagram of the display interface provided in the embodiments of this application;

[0034] Figure 5C Another schematic diagram of the display interface provided in the embodiments of this application;

[0035] Figure 5D Another schematic diagram of the display interface provided in the embodiments of this application;

[0036] Figure 5E Another schematic diagram of the display interface provided in the embodiments of this application;

[0037] Figure 5F Another schematic diagram of the display interface provided in the embodiments of this application;

[0038] Figure 5G A schematic diagram of an embodiment of the patent search method provided in this application;

[0039] Figure 6 This is a virtual structural diagram of the patent search device provided in the embodiments of this application;

[0040] Figure 7 This is a schematic diagram of the hardware structure of the server provided in an embodiment of this application. Detailed Implementation

[0041] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0042] The following explanation of the search method is from the perspective of a patent search device, which can be a server or a service unit within a server; no specific limitation is made.

[0043] Please refer to the following: Figure 1 , Figure 1 A flowchart illustrating the patent search method provided in this application embodiment includes:

[0044] 101. Provide a search interface.

[0045] In this embodiment, the patent search device can provide a search interface. Please refer to [link / reference]. Figure 2B , Figure 2B This is a schematic diagram of the search interface provided in the embodiments of this application, including a search box 201 and a search parameter area 202.

[0046] 102. Determine the operation instructions based on the user's conditions. Determine the search conditions on the search interface.

[0047] In this embodiment, the patent search device can receive a user's condition determination operation instruction and determine the search conditions on the search interface according to the condition determination operation instruction. For example... Figure 2B As shown, the user inputs a condition determination command into the search box 201 of the search interface. The patent search device can receive this condition determination command and thereby determine the search conditions. These search conditions can be at least one patent number entered by the user, such as "CN103001244A" ( Figure 2A The search box 201 allows users to input multiple patent documents, or the specific content of a single patent document, or even just one article; there are no specific limitations. Understandably, the patent search device can receive user commands, allowing users to retrieve all claims of a corresponding patent document by selecting "Get All Claims." Similarly, selecting "Inventor" and "Applicant" allows users to retrieve all other patent documents invented or applied for by the inventors and applicants of the patent document being searched.

[0048] It should be noted that after determining the search criteria, the patent search device can obtain the target dataset corresponding to those search criteria, such as... Figure 2B As shown, selection is made using 201-A, choosing the corresponding number of items. This means selecting patent documents from the database that correspond to the search criteria based on semantic relevance. For example, the top 400 patent documents in the database that match the search criteria can be used as the target dataset (of course, other numbers of items are also possible, such as the top 1000; there is no specific limitation). Alternatively, as shown... Figure 2A As shown, in 201-A, select "Import File" to import a file to determine the target dataset corresponding to the search criteria.

[0049] 103. Determine the search parameters corresponding to the search criteria.

[0050] In this embodiment, the patent search device can determine the search parameters corresponding to the search conditions. These search parameters include at least one of keywords, IPC, and CPC, such as... Figure 2C As shown, the user can operate the keyword acquisition control 203 in the search parameter area 202. The patent search device can receive the user's operation instruction on the keyword acquisition control 203, and retrieve the keywords corresponding to the patent "CN103001244A" according to the operation instruction, such as "power", "sub-processor", "input terminal" and "controller" keywords, and display the retrieved keywords, such as... Figure 2B The keyword corresponding to the obtained patent “CN103001244A” will be displayed in the keyword display area 206. At the same time, the user can also operate the IPC control 204 and the CPC control 205. The patent search device can receive the user’s operation instructions for the IPC control 204 and the CPC control 205 and obtain the IPC and CPC corresponding to the patent “CN103001244A” according to the operation instructions.

[0051] It should be noted that users can also operate the enhanced sorting control 207 and other parameter controls 208 in the search parameter area 202, thereby enhancing the sorting of search results or inputting other search parameters, including but not limited to at least one of the following parameters:

[0052] Publication (Announcement) Number, Application Number, Priority Number, First Inventor, Inventor, Applicant, First Applicant, Patentee, Patent Licensor, Patent Pledgee, Patent Assignor, Patent Assignee, Patent Citation, Patent Citation Applicant, Patent Citation Applicant, Patent Family, Simple Patent Family, Extended Patent Family, Patent Family Applicant, Patent Family Inventor, Application Date, Publication Date, Grant Date, Priority Date, Patent Expiry Date, Patent Rejection Date, Patent Withdrawal Date, Patent Pledge Date, Patent License Date, Patent Pledge Date, Patent Assignment Date, Patent Assignee Date, Patent License Date, Patent Preservation Date, Patent Reexamination and Invalidation Decision Date, Patent Type, Patent Applicant's Geographical Location, Patent Legal Status, Patent Full Text Data The data includes existing patent search system search fields and user-defined data such as patent title text data, patent abstract text data, patent description text data, number of patent rights, number of patent technical features, patent novelty score, patent value, number of patent citations, number of patent citations received, number of patent citation applicants, number of patent citation countries, number of patent citation applicants, number of patent citation applicant countries, patent agency, patent attorney, patent application type, number of patent applicants, number of first patent applicants, number of patent standard applicants, number of patent inventors, number of first inventors, patent assignors, patent standard assignors, patent pledgors, patent standard pledgors, patent pledgee litigation, invalidation, success or failure information, and patent expiration status.

[0053] In one embodiment, the patent search device also performs the following operations:

[0054] Receive adjustment instructions from users;

[0055] Adjust the search parameters according to the adjustment operation instructions.

[0056] In this embodiment, the patent search device can also receive adjustment operation instructions from the user and adjust the search parameters according to the adjustment operation instructions, such as... Figure 2D As shown, users can operate on the keywords in the keyword display area 206. The patent search device can receive the operation instruction and select or deselect the keywords in the keyword display area 206 according to the operation instruction. At the same time, users can also operate on the automatic keyword 209. The patent search device can receive the operation instruction and determine the newly added keyword according to the operation instruction.

[0057] It should be noted that users can also operate on the database selection area 210 in the search interface. The patent search device can receive the user's operation instructions on the database selection area 210, and then select the corresponding database according to the operation instructions. For example, the user can add "US (US Patent Database)" as needed, where the default selected database is "CN (Chinese Patent Database)". After determining the database according to the user's selection, the patent search device can determine the target dataset corresponding to the search conditions from the database.

[0058] It should be noted that users can also operate on area 211. The patent search device can select or deselect "technical route", "cube" and "hit count sorting" according to the operation instructions. Among them, "technical route" includes 8 technical topic groups, each of which includes multiple patent documents. At the same time, the legal status of the patent documents in the database to be searched can also be filtered according to the operation instructions. For example, only patent documents with a legal status of valid can be selected, only patent documents with a legal status of published can be selected, or all patent documents can be selected. There are no specific limitations.

[0059] It should also be noted that after determining all the search parameters corresponding to the search criteria, the patent search device can receive a save command from the user, such as... Figure 2D As shown, when a user operates the save control 212, the patent search device can receive the user's save instruction and store all currently configured search parameters as a search file according to the save instruction for the user's future use. Similarly, when a user operates the load control 213, the patent search device can receive the user's load instruction, import the pre-configured search file according to the load instruction, and perform a search based on the search parameters corresponding to the pre-configured search file.

[0060] 104. Determine the P combinations of search parameters corresponding to the search parameters.

[0061] In this embodiment, after determining the search parameters corresponding to the search conditions, the patent search device can arbitrarily combine at least two of the keywords, IPC, and CPC in the search parameters to obtain P search parameter combinations, where N is an integer greater than or equal to 1. For ease of understanding, the following explanation uses keywords in the search parameters as an example to illustrate arbitrary combinations (of course, keywords can also be combined with IPC, or keywords with CPC, and the specific combination is not limited). For example, if the keywords in the search parameters include "power," "sub-processor," and "input terminal," the patent search device can arbitrarily combine these three keywords to obtain P search parameter combinations, specifically:

[0062] Search for parameter combinations such as "Power", "Sub-point", and "Input Terminal";

[0063] Search for the combination of "power" and "sub-location" parameters;

[0064] Search for parameter combinations using "Power" and "Input".

[0065] "Sub-location" and "Input end" are search parameters;

[0066] "Power" search parameter combination;

[0067] "Sub-location" search parameter combinations;

[0068] The "input end" searches for a combination of parameters.

[0069] It should be noted that the above examples of combining search parameters are just illustrations. Of course, there are other ways to combine them, such as calculating the relevance of each keyword (this relevance refers to the degree of association between two keywords, for example, "power" and "power conversion" have a high degree of association), combining keywords based on similarity, and combining synonyms, near-synonyms, etc. There are no specific limitations, as long as the keywords, IPC, and CPC in the search parameters can be combined arbitrarily.

[0070] 105. Search the target dataset based on P combinations of search parameters to obtain the patent documents matched by each of the P combinations of search parameters.

[0071] In this embodiment, after determining P combinations of search parameters, the patent search device can search the target dataset according to the P combinations of search parameters to obtain the patent documents matched by each of the P combinations of search parameters. The target dataset is the dataset selected by the patent search device from the database according to the search conditions.

[0072] In one embodiment, the patent search device also performs the following operations:

[0073] Provide a display interface;

[0074] The target dataset is sorted and displayed on the display interface according to the number of hit search parameters or relevance;

[0075] The patent document set matched by each of the P search parameter combinations is deduplicated so that the patent documents only appear under the search parameter combination node with the most matched search parameters. Each search parameter combination node includes M patent documents, where M is an integer greater than or equal to 0.

[0076] The collection of patent documents matched by each of the P unique search parameter combinations is sorted and displayed on the display interface according to the number of matched search parameters or relevance.

[0077] In this embodiment, the patent search device can provide a display interface and sort the target dataset on the display interface according to relevance. That is, after selecting the target dataset corresponding to the search conditions from the database according to the search conditions, the target dataset can be sorted according to relevance. In addition, after retrieving the patent document set matched by each of the P search parameter combinations, the patent search device can also deduplicate the patent document set matched by each of the P search parameter combinations, so that the patent documents only appear under the search parameter combination node with the most matched search parameters. Each search parameter combination node includes M patent documents, where M is an integer greater than or equal to 0. The deduplicated patent document set is then sorted and displayed on the display interface according to the number of matched search parameters or relevance (that is, by selecting the "cube" in the search interface in Figure 2, the patent documents corresponding to each search parameter combination are deduplicated, and it is ensured that the patent documents only appear under the search parameter combination node with the most matched keywords, and only once). For ease of understanding, the following is combined with Figures 3A to 3D The following example illustrates how arbitrary combinations of keywords in search parameters can be used to obtain a set of patent documents that are matched by that combination:

[0078] Please see Figure 3A , Figure 3A A schematic diagram of the display interface provided in the embodiments of this application, such as Figure 3A As shown, 300 is the display interface, and 301 is the search result corresponding to the search parameters obtained by arbitrarily combining the keywords in the search parameters.

[0079] Taking the search criteria "CN103001244A" with keywords including "distributed|solar|DC power supply|power|power conversion|power converter|controller|temperature sensor|temperature signal|input power input terminal|input terminal|output power|output terminal|output terminal|" as an example, where "|" is the separator between keywords, the patent search device can combine these keyword combinations to obtain multiple corresponding search parameter combinations, and then search the target dataset according to the search parameter combinations to obtain the patent documents (such as...) matched by each search parameter combination. Figure 3D As shown, Figure 3DThe diagram illustrates the matching of search parameter combinations with the patent document sets they match. For example, the search parameter combination "DC power supply | power | power conversion | power converter | input terminal | output terminal" is obtained by combining keywords from the search parameters. The patent document sets matched by this search parameter combination are "CN102510218A" and "CN1897402A". Then, the number of keywords in each search parameter combination is determined, and the patent documents matched by each search parameter combination are sorted and displayed according to the number of keywords in each combination. Figure 3A As shown, the patent search device can sort the search parameter combinations according to the number of keywords hit, and place the patent documents hit by each search parameter combination under the corresponding search parameter combination node.

[0080] It should be noted that the patent search device can also perform searches based on the "technical route," "hit count sorting," IPC, and CPC settings on the search interface shown in Figure 2, and obtain the corresponding search results, such as... Figure 3B As shown, 302 represents the search results corresponding to "Hit Count Sorting", which is the improved semantic relevance sorting of the Chinese patent application database. The number in front of each patent document is the Boolean hit count. 303 represents the search results corresponding to "Technical Route". 304 represents the search results corresponding to CPC. Each CPC search result node includes multiple patent documents. 305 represents the search results corresponding to IPC.

[0081] In addition, none of the patent documents included in "counterexample" 305-A contain the keywords in the search parameters. Figure 3B The "second-order counterexample" 306 is a search result obtained by searching for "CN103001244A" based on semantic search or other search conditions. Furthermore, the patent documents included in this "second-order counterexample" do not contain the keywords, IPC, and CPC from the search parameters. Thus, the search results include not only the matched combinations but also documents that were not matched and may have been missed, ensuring a comprehensive and accurate search.

[0082] After obtaining the patent documents corresponding to the search criteria, the patent search device can receive user operation instructions and display the search information of the selected patent documents according to the operation instructions, such as... Figure 3CAs shown, 306-A represents the search information for the selected patent document. This search information includes, but is not limited to, patent number, patent value, patent quality, application number, patent type, title, application date, publication date, priority date, applicant, inventor, international classification, CPC classification, citation information, cited information, family, claims, number of classifications, legal status, relevance (which refers to the correspondence with the search conditions), and the node where the patent document is located. The patent document selected by 306-A is a patent document under the "second-order counterexample" node, and its relevance to the patent document with application number CN103001244A is 92.

[0083] It should be noted that after sorting and displaying the patent documents corresponding to each combination of search parameters on the display interface, the patent search device also performs the following operations:

[0084] The interface receives the user's search alignment instructions.

[0085] The search parameters matched by each search parameter combination node in the P search parameter combinations are aligned according to the search alignment instruction. Each search parameter combination node includes M patent documents, where M is an integer greater than or equal to 0.

[0086] like Figure 4A As shown, Figure 4A The following explanation uses the combination of search parameters corresponding to keywords as an example. 401 is the result of aligning the keywords in the search parameter combination according to the search alignment instruction, that is, aligning the same keywords contained in each search parameter combination.

[0087] In addition, the patent search device can also receive a first aggregation instruction from the user on the display interface; determine a first search parameter based on the first aggregation instruction; and aggregate each of the P search parameter combinations according to the first aggregation parameter to obtain a first aggregation result. For example... Figure 4B As shown, 402 is a schematic diagram of the first aggregation result. Taking the aggregation of the keyword "distributed" as an example, it can also be used to perform aggregation operations on other keywords.

[0088] It should also be noted that the patent search device can further aggregate the results based on the first aggregation result. Specifically: the patent search device receives a second aggregation instruction from the user on the display interface; determines second search parameters based on the second aggregation instruction; and aggregates the patent document set based on the second search parameters and the first aggregation result to obtain the second aggregation result. For example... Figure 4CAs shown in Figure 403, this is a schematic diagram of the second aggregation result. The second aggregation result is obtained by performing a secondary aggregation on the first aggregation result using the keyword "solar energy". For example, search parameter combinations that simultaneously satisfy the first aggregation parameter "distributed" and the second aggregation parameter "solar energy" are aggregated first.

[0089] It should also be noted that the patent search device can also determine a target search parameter combination based on the user's search parameter input instruction. The target search parameter combination includes at least one search parameter. The target search parameter combination is matched with the P search parameter combinations to obtain at least two sets of search parameter combinations, wherein the at least two sets of search parameter combinations cover the most search parameters in the target search parameter combination.

[0090] Furthermore, the patent search device can also receive a target search parameter combination input by the user and match this target search parameter combination with P search parameter combinations to obtain at least two sets of search parameter combinations corresponding to the target search parameter combination. These at least two sets of search parameter combinations match the target search parameter combination with the most search parameters. The at least two sets of search parameter combinations are matched with the target search parameter combination through union or optimal matching distance, where optimal matching distance minimizes the number of search parameter combinations required to cover the target search parameter combination. Thus, the set of patent documents matched by the target search parameter combination can be obtained.

[0091] In one embodiment, the patent search device also performs the following operations:

[0092] The interface receives targeted supplementary search instructions from users.

[0093] Select the target patent document according to the targeted supplementary search instructions;

[0094] Identify the supplementary set of patent documents corresponding to the target patent document.

[0095] The following is combined Figures 5A to 5B To explain, in the display interface 500, the patent search device can receive the user's targeted supplementary search instruction and determine the target patent document based on the instruction. For example, as shown in 501, "CN101087069A" is used as an example. The device then determines the supplementary patent document set corresponding to this target patent document. "CN101087069A" has a relevance of 81. The user considers this patent important and can use it as a basis for targeted supplementary searching, which means lowering the relevance of the current document by L, where L is a relevance of 1-99%. For example, lowering the relevance by 5% means selecting patents in the target dataset with a relevance of 76%-81% to the search criteria as the supplementary patent document set and displaying them. Figure 5B As shown in 502.

[0096] In one embodiment, the patent search device also performs the following operations:

[0097] Obtain the first semantic ranking relevance corresponding to the target dataset;

[0098] Receive the user's enhanced sorting instruction;

[0099] The second semantic ranking relevance of the target dataset is calculated according to the enhanced ranking instruction, wherein the second semantic ranking relevance has at least one dimension.

[0100] The target dataset is sorted according to the first semantic relevance and the second semantic ranking relevance.

[0101] The following is combined Figures 5C to 5F Explaining augmented sorting, Figure 5C The value 503 in the diagram shows the ranking result of the patent document set before enhanced ranking. That is, the patent document set is ranked according to the first semantic relevance corresponding to the target dataset. Then, the patent search device can receive the user's enhanced ranking instruction and calculate the second semantic ranking relevance corresponding to the target dataset based on the instruction. This second semantic ranking relevance has at least one dimension. The target dataset is then ranked based on both the first and second semantic ranking relevance. For example... Figure 5D As shown in section 504, when the "enhanced sorting" function is invoked, the patent search device can display an enhanced sorting settings window based on this enhanced sorting instruction, such as... Figure 5E The enhanced sorting settings window 505 shown allows users to set the features corresponding to the enhanced sorting. For example, a certain claim item with the publication number "CN103001244A" can be loaded as the feature corresponding to the enhanced sorting (it is understood that the feature of the enhanced sorting can also be other features, such as abstract, background technology, use, beneficial effects, prior art, or a paragraph entered by the user, without any specific limitation).

[0102] It should be noted that users can also, of course, [do something else]. Figure 2C The feature corresponding to the enhanced ranking is set in the middle. If in Figure 2C The system includes settings for enhanced sorting. Simply select enhanced sorting and then perform a search to perform enhanced sorting based on the existing search results.

[0103] like Figure 5FAs shown, any patent document in the enhanced ranking set includes two relevance scores: one is the relevance score corresponding to semantic retrieval, and the other is the enhanced relevance score related to the feature corresponding to "enhanced ranking". Patent documents with the same enhanced relevance score are also grouped and ranked. For example, the enhanced relevance scores of all patent documents in set 506 are 85.

[0104] The following is a simplified example: See Figure 5G .

[0105] "R / cn1234" represents the first semantic ranking, meaning the first semantic ranking benchmark is cn1234. When only the first semantic ranking is considered, if the search result set has 5 elements {cn12, cn23, cn34, cn56, cn67}, and the relevance of each search result to cn1234 is {0.83, 0.84, 0.85, 0.85, 0.85} respectively, the first semantic ranking result would be {cn34, cn67, cn56, cn23, cn12}.

[0106] Considering only the second semantic ordering,

[0107] R / “Example…”, meaning the second semantic ranking benchmark is a piece of text, assuming the relevance between the search result set and the second semantic ranking benchmark is {0.92, 0.60, 0.90, 0.60, 0.92}, the second semantic ranking result is {cn12,cn67,cn34,cn56,cn23}. In existing technology, the ranking result can only consider the last ranking, such as the first ranking R / cn1234 and the second ranking R / “Example…”, the final ranking result is determined by the second ranking R / “Example…”.

[0108] The "enhanced ordering" in this application considers both first semantic ordering and second semantic ordering:

[0109] Suppose the search result set has 5 elements {cn12,cn23,cn34,cn56,cn67}. The relevance of each search result to the first semantic ranking benchmark cn1234 and the second semantic ranking benchmark "Example..." in n dimensions (here, 2 dimensions, but not limited to 2 dimensions) is represented as {(0.83,0.92), (0.84,0.60), (0.85,0.90), (0.85,0.60),(0.85,0.92)}.

[0110] The "enhanced ranking" in this application extracts the relevance of any one dimension from any other dimension as the ranking relevance. For illustrative purposes, this embodiment extracts the second ranking relevance as the ranking relevance. The second ranking relevance of cn12 is 0.92, which is greater than that of cn23 (0.6), cn34 (0.9), and cn56 (0.6), but the second ranking relevance of cn67 (0.92) is also 0.92, the same as that of cn12. Existing technology uses only one-dimensional relevance values ​​to determine ranking, which obviously cannot obtain other relevance information to determine the ranking of two elements with the same ranking relevance.

[0111] The "enhanced ranking" method employed in this application involves using a different ranking relevance (e.g., but not limited to, the first ranking relevance) to determine the ranking value of two elements when two elements have the same second ranking relevance. In this embodiment, the second ranking relevance of both cn12 and cn67 is 0.92. Further comparison of the first ranking relevances of the two elements reveals that cn12 has 0.83 and cn67 has 0.92. Since cn67's first relevance of 0.92 is greater than cn12's first relevance of 0.83, cn67 ranks before cn12.

[0112] Thus, through the "enhanced sorting" of this application, the sorting of the retrieval result set is as follows:

[0113] (R / cn1234, R / “Example…”) = {cn67,cn12,cn34,cn56,cn23}

[0114] Furthermore, this application displays the multi-dimensional semantic ranking results using color modulation. Color modulation displays documents with the same second ranking relevance as shown in the same color. For example, in this embodiment, cn67 and cn12 have the same second ranking relevance value of 0.92, and the background color (or other representational features) of these two documents is displayed in the same or different colors through color modulation; in this embodiment, a yellow background color is used. Furthermore, colors can be modulated to display documents with the same ranking relevance. For example, in this embodiment, cn34 has an independent second ranking relevance of 0.90, and is represented by a different background color; in this embodiment, white is used. Furthermore, cn56 and cn23 have the same second ranking relevance of 0.6, and are displayed with a yellow background color through color modulation.

[0115] The enhanced sorting and color modulation of this invention can facilitate the discovery of the semantic distribution of block-like and strip-like structures in high-dimensional semantic sorting.

[0116] In summary, it can be seen that in the embodiments provided in this application, the patent search device determines the search conditions on the search interface based on the user's conditions and operation instructions; determines the search parameters corresponding to the search conditions, including keywords, IPC, and CPC; arbitrarily combines the keywords, IPC, and CPC in the search parameters to obtain P search parameter combinations; and searches the target dataset based on the P search parameter combinations to obtain the patent documents matched by each of the P search parameter combinations. Therefore, the user does not need to read every patent document retrieved; they can find the most suitable search parameter combination by browsing the search parameter combination tags, thereby retrieving the corresponding patent documents, greatly reducing search time.

[0117] The above describes this application from the perspective of patent search methods; the following describes this application from the perspective of patent search devices.

[0118] Please see Figure 6 , Figure 6 This is a virtual structural diagram of a patent search device provided in an embodiment of this application. The patent search device 600 includes:

[0119] The interface providing unit 601 is used to provide a search interface;

[0120] The first determining unit 602 is used to determine the search conditions on the search interface based on the user's conditions and operation instructions.

[0121] The second determining unit 603 is used to determine the search parameters corresponding to the search conditions, wherein the search parameters include at least one of keywords, IPC, and CPC;

[0122] Combination unit 604 is used to determine P combinations of search parameters corresponding to the search parameters, wherein N is an integer greater than or equal to 1;

[0123] The retrieval unit 605 is used to retrieve the target dataset based on the P retrieval parameter combinations to obtain the set of patent documents matched by each of the P retrieval parameter combinations, with the target data corresponding to the retrieval conditions.

[0124] In one possible design, the device further includes:

[0125] Display unit 606, the display unit 606 is used for:

[0126] Provide a display interface;

[0127] The target dataset is sorted and displayed on the display interface according to the number of hit search parameters or relevance;

[0128] The patent document set matched by each of the P search parameter combinations is deduplicated so that the patent documents only appear under the search parameter combination node with the most matched search parameters. Each search parameter combination node includes M patent documents, where M is an integer greater than or equal to 0.

[0129] The collection of patent documents matched by each of the P unique search parameter combinations is sorted and displayed on the display interface according to the number of matched search parameters or relevance.

[0130] In one possible design, the retrieval unit 605 is further configured to:

[0131] The user's search alignment command is received on the display interface;

[0132] The search parameters matched by each search parameter combination node in the P search parameter combinations are aligned according to the search alignment instruction.

[0133] In one possible design, the retrieval unit 605 is further configured to:

[0134] The user's first aggregation command is received on the display interface;

[0135] The first search parameters are determined according to the first aggregation instruction;

[0136] The patent document set is aggregated according to the first search parameters to obtain the first aggregation result.

[0137] In one possible design, the retrieval unit 605 is further configured to:

[0138] The second aggregation command from the user is received on the display interface;

[0139] The second search parameters are determined according to the second aggregation instruction;

[0140] The patent document set is aggregated based on the second search parameters and the first aggregation result to obtain a second aggregation result.

[0141] In one possible design, the retrieval unit 605 is further configured to:

[0142] The target search parameter combination is determined based on the user's search parameter input instruction, and the target search parameter combination includes at least one search parameter;

[0143] The target retrieval parameter combination is matched with the P retrieval parameter combinations to obtain at least one set of retrieval parameter combinations, wherein the at least one set of retrieval parameter combinations covers all retrieval parameters in the target retrieval parameter combination.

[0144] In one possible design, the retrieval unit 605 is further configured to:

[0145] The system receives the user's targeted supplementary search instruction on the display interface.

[0146] Select the target patent document according to the targeted supplementary search instructions;

[0147] Determine the supplementary set of patent documents corresponding to the target patent document.

[0148] In one possible design, the retrieval unit 605 is further configured to:

[0149] Receive the user's save command;

[0150] According to the save instruction, all currently configured search parameters are stored as a search file;

[0151] or,

[0152] Receive the user's loading instruction;

[0153] Import the pre-configured search file according to the loading instruction;

[0154] The search is performed based on the search parameters corresponding to the pre-configured search file.

[0155] In one embodiment, the retrieval unit 605 is further configured to:

[0156] Obtain the first semantic ranking relevance corresponding to the target dataset;

[0157] Receive the user's enhanced sorting instruction;

[0158] The second semantic ranking relevance of the target dataset is calculated according to the enhanced ranking instruction, wherein the second semantic ranking relevance has at least one dimension.

[0159] The target dataset is sorted according to the first semantic relevance and the second semantic ranking relevance.

[0160] Figure 7 This is a schematic diagram of the server structure in this application, as shown below. Figure 7As shown, the server 700 in this embodiment includes at least one processor 701, at least one network interface 704 or other user interface 703, a memory 705, and at least one communication bus 702. The server 700 may optionally include the user interface 703, including a display, keyboard, or clicking device. The memory 705 may include high-speed RAM, or it may also include non-volatile memory, such as at least one disk storage device. The memory 705 stores execution instructions. When the server 700 is running, the processor 701 communicates with the memory 705, and the processor 701 calls the instructions stored in the memory 705 to execute the aforementioned retrieval method. The operating system 706 contains various programs for implementing various basic services and processing tasks based on the hardware.

[0161] The server provided in this application embodiment can execute the technical solution of the above-described patent search method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.

[0162] This application also provides a computer-readable storage medium storing a computer program thereon. When executed by a computer, the computer program implements the method flow related to the patent search device in any of the above method embodiments. Correspondingly, the computer can be the aforementioned patent search device.

[0163] This application also provides a computer program or a computer program product including a computer program, which, when executed on a computer, causes the computer to implement the method flow related to the patent search device in any of the above method embodiments. Correspondingly, the computer can be the aforementioned patent search device.

[0164] In the above Figure 1 In the illustrated embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0165] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A patent search method, characterized in that, include: Provide a search interface; Based on the user's criteria, the operation instructions are determined on the search interface, specifying the search conditions. Determine the search parameters corresponding to the search criteria, wherein the search parameters include at least one of keywords, IPC, and CPC; Determine P combinations of search parameters corresponding to the search parameters, where P is an integer greater than or equal to 1; The target dataset is retrieved based on the P search parameter combinations to obtain the set of patent documents matched by each of the P search parameter combinations. The target dataset is the dataset selected from the database that corresponds to the search conditions based on the search conditions. Provide a display interface; Displays the set of patent documents matched by the combination of search parameters with parameter labels.

2. The method according to claim 1, characterized in that, The method further includes: The target dataset is sorted and displayed on the display interface according to the number of hit search parameters or relevance; The patent document sets matched by each of the P search parameter combinations are deduplicated so that the patent documents only appear under the search parameter combination node with the most matched search parameters. Each search parameter combination node includes M patent documents, where M is an integer greater than or equal to 0. The collection of patent documents matched by each of the P search parameter combinations after deduplication is displayed on the display interface in sorted order by the number of matched search parameters or relevance.

3. The method according to claim 2, characterized in that, The method further includes: The user's search alignment command is received on the display interface; The search parameters matched by each search parameter combination node in the P search parameter combinations are aligned according to the search alignment instruction.

4. The method according to claim 3, characterized in that, The method further includes: The user's first aggregation command is received on the display interface; The first search parameters are determined according to the first aggregation instruction; The patent document set is aggregated according to the first search parameters to obtain the first aggregation result.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: The target search parameter combination is determined based on the user's search parameter input instruction, and the target search parameter combination includes at least one search parameter; The target retrieval parameter combination is matched with the P retrieval parameter combinations to obtain at least two sets of retrieval parameter combinations, wherein the at least two sets of retrieval parameter combinations cover the most retrieval parameters in the target retrieval parameter combination.

6. The method according to any one of claims 1 to 2, characterized in that, The method further includes: The system receives the user's targeted supplementary search instruction on the display interface. Select the target patent document according to the targeted supplementary search instructions; Determine the supplementary set of patent documents corresponding to the target patent document.

7. The method according to claim 1, characterized in that, The method further includes: Receive the user's save command; According to the save instruction, all currently configured search parameters are stored as a search file; or, Receive the user's loading instruction; Import the pre-configured search file according to the loading instruction; The search is performed based on the search parameters corresponding to the pre-configured search file.

8. The method according to any one of claims 1 to 2, characterized in that, The method further includes: Obtain the first semantic ranking relevance corresponding to the target dataset; Receive the user's enhanced sorting instruction; The second semantic ranking relevance of the target dataset is calculated according to the enhanced ranking instruction, wherein the second semantic ranking relevance has at least one dimension. The target dataset is sorted according to the first semantic relevance and the second semantic ranking relevance.

9. A patent search device, characterized in that, include: The interface providing unit is used to provide a search interface; The first determining unit is used to determine the operation instructions based on the user's conditions and to determine the search conditions on the search interface. The second determining unit is used to determine the search parameters corresponding to the search conditions, wherein the search parameters include at least one of keywords, IPC, and CPC; A combination unit is used to determine P combinations of search parameters corresponding to the search parameters, wherein P is an integer greater than or equal to 1; The retrieval unit is used to retrieve the target dataset based on the P retrieval parameter combinations to obtain the set of patent documents matched by each of the P retrieval parameter combinations. The target dataset is a dataset selected from the database that corresponds to the retrieval conditions based on the retrieval conditions. The display unit provides a display interface that shows the set of patent documents matched by the combination of search parameters with parameter labels.

10. A computer storage medium, characterized in that, include: Instructions, when executed on a computer, cause the computer to perform the steps of the patent search method according to any one of claims 1 to 8.

Images