A search optimization method, device, equipment and storage medium thereof

Abstract

The embodiment of the application belongs to the technical field of financial science and technology, is applied to the optimization scene of a new search channel of a financial service, and relates to a search optimization method, device and equipment and a storage medium thereof, comprising the following steps: sending a search field to a plurality of preset search channels for data search, wherein the plurality of search channels comprise an old search channel and a new search channel; identifying a named entity in the search field and a search intention corresponding to a data search request according to a front-end processing component in the old search channel; acquiring search recall results of the plurality of search channels respectively; and performing search optimization on the new search channel based on the search recall results, the named entity, the search intention and a preset search optimization strategy. The new search channel is conveniently optimized in combination with the old search channel, the search optimization of the new search channel is ensured to combine the relatively mature old search channel, and the search recall result accuracy of the new search channel after the search optimization is indirectly ensured.

Description

Technical Field

[0001] This application relates to the field of financial technology and is applied to the optimization of new search channels in financial business. In particular, it relates to a search optimization method, apparatus, device and its storage medium. Background Technology

[0002] With the rapid development of the internet, all industries are seeking breakthroughs by leveraging it. In recent years, the financial industry has also been expanding its online business. Due to the large volume of business and data involved in the financial industry, there is also a significant demand for financial data search services.

[0003] Currently, when conducting financial data searches, enterprises need to connect to multiple channels. A single search provider may need to offer search services to multiple similar channels, exhibiting high similarity in users and search content, but the content itself is not entirely identical. Internal or external search engines often need to connect to internal or external channel content providers. A common need is that some older search channels have large traffic, abundant data, and numerous features, making it easier to develop and train models. However, some newer search channels have less data and fewer users, making it difficult for search recall and ranking models to obtain data and perform feature processing. The limited data on newer search channels also leads to lower accuracy in search recall results. Summary of the Invention

[0004] The purpose of this application is to provide a search optimization method, apparatus, device and storage medium to solve the problem that the accuracy of search recall results of new search channels is low when using them in the prior art, due to the limited data and users of the new search channels.

[0005] To address the aforementioned technical problems, this application provides a search optimization method, which employs the following technical solution:

[0006] A search optimization method includes the following steps:

[0007] Step 201: Obtain the data search request and search fields sent by the target search terminal;

[0008] Step 202: In response to the data search request, the search field is sent to several preset search channels for data search, wherein the several search channels include old search channels and new search channels;

[0009] Step 203: Based on the pre-processing component in the old search channel, identify the named entities in the search field and the search intent corresponding to the data search request;

[0010] Step 204: Obtain the search recall results for the aforementioned search channels respectively;

[0011] Step 205: Based on the search recall results, the named entity, the search intent, and the preset search optimization strategy, optimize the new search channel.

[0012] Furthermore, the pre-processing component includes a word segmentation sub-component, a named entity recognition sub-component, and an intent recognition sub-component. The step of identifying the named entities in the search field and the search intent corresponding to the data search request based on the pre-processing component in the old search channel specifically includes:

[0013] The search field is input into the word segmentation sub-component to obtain the word segmentation result output by the word segmentation sub-component, wherein the word segmentation sub-component includes the jieba word segmentation plugin;

[0014] The word segmentation result is input into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent;

[0015] The search field and the data search request are input into the intent recognition subcomponent. The search intent corresponding to the data search request is determined based on the intent recognition subcomponent. The intent recognition subcomponent is a semantic intent recognition plugin based on NLP.

[0016] Furthermore, the step of obtaining the search retrieval results from the plurality of search channels specifically includes:

[0017] Based on the search fields and the search intent, construct SQL execution statements adapted to the various search channels respectively;

[0018] Based on the SQL execution statements adapted to the various search channels and the search functions built into the various search channels, search processing is performed to obtain the search recall results of the various search channels.

[0019] After performing the step of obtaining the search recall results of the plurality of search channels respectively, the method further includes:

[0020] Obtain the distinguishing identifiers of the aforementioned search channels;

[0021] The distinguishing identifier information is assigned to the search recall results corresponding to the various search channels.

[0022] Furthermore, the step of optimizing the new search channel based on the search recall results, the named entity, the search intent, and the preset search optimization strategy specifically includes:

[0023] Step 401: Using a preset similarity algorithm and the distinguishing identifier information, calculate the similarity between the search recall results of the new search channel and the search recall results of all old search channels.

[0024] Step 402: By cyclic comparison, determine whether the similarity between the search recall results of the current old search channel and the search recall results of the new search channel meets the preset similarity threshold.

[0025] Step 403: If the similarity between the search recall result of the current old search channel and the search recall result of the new search channel meets the similarity threshold, then update the next old search channel to the current old search channel and execute step 402 until all old search channels have been compared with the new search channel, and stop executing step 402.

[0026] Step 404: If the similarity between the search recall results of the current old search channel and the search recall results of the new search channel does not meet the similarity threshold, then the new search channel is optimized based on the search recall results, named entities, and search intent corresponding to the current old search channel.

[0027] Furthermore, after performing the step of optimizing the new search channel based on the search recall results, named entities, and search intent corresponding to the current old search channel, the method further includes:

[0028] Step 501: Resend the search field to the new search channel for data search and obtain the latest search recall results corresponding to the new search channel;

[0029] Step 502: Calculate the similarity between the latest search recall result and the search recall results of all old search channels using the similarity algorithm and the distinguishing identifier information.

[0030] Step 503: By comparison, determine whether the similarity between the search recall results of the current old search channel and the latest search recall results meets the similarity threshold.

[0031] Step 504: If the similarity between the search recall result of the current old search channel and the latest search recall result does not meet the similarity threshold, then continue to optimize the search for the new search channel based on the search recall result, named entity and search intent corresponding to the current old search channel.

[0032] Step 505: Repeat steps 501 to 503 until the similarity between the search recall results of all old search channels and the latest search recall results corresponding to the new search channel meets the similarity threshold. Then stop repeating steps 501 to 503 to complete the search optimization for the new search channel.

[0033] Furthermore, the step of optimizing the new search channel based on the search recall results, named entities, and search intent corresponding to the current old search channel specifically includes:

[0034] The named entities are transferred as search entity tags to the new search channel using transfer learning.

[0035] The search intent is migrated to the new search channel, and the search intent is set as the search intent of the new search channel;

[0036] The search recall results are migrated to a pre-set search result cache library in the new search channel, and a knowledge mapping relationship is constructed between the named entity, the search intent, and the search recall results within the new search channel.

[0037] To address the aforementioned technical problems, embodiments of this application also provide a search optimization device, which employs the following technical solution:

[0038] A search optimization device, comprising:

[0039] The search information acquisition module is used to acquire the data search request and search fields sent by the target search terminal.

[0040] A data search module is used to respond to the data search request and send the search field to a number of preset search channels for data search, wherein the number of search channels includes old search channels and new search channels;

[0041] The pre-processing module is used to identify the named entities in the search field and the search intent corresponding to the data search request based on the pre-processing components in the old search channel.

[0042] The search recall result acquisition module is used to acquire the search recall results of the plurality of search channels respectively;

[0043] The new search channel search optimization module is used to optimize the new search channel based on the search recall results, the named entities, the search intent, and the preset search optimization strategy.

[0044] Furthermore, the pre-processing module includes:

[0045] The word segmentation processing submodule is used to input the search field into the word segmentation subcomponent and obtain the word segmentation result output by the word segmentation subcomponent, wherein the word segmentation subcomponent includes the jieba word segmentation plugin;

[0046] The named entity recognition submodule is used to input the word segmentation result into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent;

[0047] The search intent recognition submodule is used to input the search field and the data search request into the intent recognition subcomponent, and determine the search intent corresponding to the data search request based on the intent recognition subcomponent. The intent recognition subcomponent is a semantic intent recognition plugin based on NLP.

[0048] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0049] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the search optimization method described above.

[0050] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0051] A computer-readable storage medium storing computer-readable instructions that, when executed by a processor, implement the steps of the search optimization method described above.

[0052] Compared with the prior art, the embodiments of this application have the following main advantages:

[0053] The search optimization method described in this application involves: acquiring a data search request and a search field sent by a target search terminal; responding to the data search request and sending the search field to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; identifying named entities in the search field and the search intent corresponding to the data search request based on a pre-processing component in the old search channel; acquiring search recall results for each of the several search channels; and optimizing the search for the new search channel based on the search recall results, the named entities, the search intent, and a preset search optimization strategy. The pre-processing component in the old search channel processes the search field and the data search request to obtain the named font in the search field and the search intent corresponding to the data search request, facilitating subsequent search optimization for the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. By combining existing search channels with new search channels for search optimization, the optimization of new search channels is ensured to be integrated with the relatively mature existing search channels, which also indirectly ensures the accuracy of search recall results after the optimization of new search channels. Attached Figure Description

[0054] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0056] Figure 2 This is a flowchart of an embodiment of the search optimization method according to this application;

[0057] Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 203 shown;

[0058] Figure 4 yes Figure 2 A flowchart of a specific embodiment of step 205 shown;

[0059] Figure 5 yes Figure 4 A flowchart of a specific embodiment of the iterative optimization steps following step 404 is shown;

[0060] Figure 6This is a schematic diagram of one embodiment of the search optimization apparatus according to this application;

[0061] Figure 7 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0063] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0064] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0065] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0066] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0067] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 players (Moving Picture Experts Group Audio Layer IV), laptops, and desktop computers, etc.

[0068] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0069] It should be noted that the search optimization method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the search optimization device is generally set in the server / terminal device.

[0070] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0071] Continue to refer to Figure 2 The diagram illustrates a flowchart of an embodiment of the search optimization method according to this application. The search optimization method includes the following steps:

[0072] Step 201: Obtain the data search request and search fields sent by the target search terminal.

[0073] In this embodiment, the target search terminal includes a data search terminal in a financial business system, such as a client-side data search terminal or a server-side data search terminal in a financial business system.

[0074] In this embodiment, the step of obtaining the data search request and search field sent by the target search terminal specifically includes: obtaining the data search request and search field pushed by the target search terminal using a message acquisition method, wherein the search field is obtained from the search box of the target search terminal, and the data search request is generated by the target search terminal in combination with the search field.

[0075] Step 202: In response to the data search request, the search field is sent to several preset search channels for data search, wherein the several search channels include old search channels and new search channels.

[0076] In this embodiment, the "several search channels" refer to different search channels when searching for financial business data, such as self-built search libraries, financial analysis cloud platforms, etc. The "several search channels" include old search channels, that is, search channels used in the past, which are generally mature and have been verified over a long period of time. They also include new search channels, that is, newly added search channels, which are generally search channels whose search results have not yet been verified over a long period of time.

[0077] In this embodiment, the step of sending the search field to several preset search channels for data searching specifically includes: identifying the distinguishing identifiers of the several search channels; and sending the search field to each of the several search channels according to the distinguishing identifiers for data searching.

[0078] Step 203: Based on the pre-processing components in the old search channel, identify the named entities in the search field and the search intent corresponding to the data search request.

[0079] In this embodiment, the preprocessing component includes a word segmentation subcomponent, a named entity recognition subcomponent, and an intent recognition subcomponent.

[0080] Continue to refer to Figure 3 , Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 203 shown includes:

[0081] Step 301: Input the search field into the word segmentation sub-component to obtain the word segmentation result output by the word segmentation sub-component, wherein the word segmentation sub-component includes the JIEBA word segmentation plugin;

[0082] Step 302: Input the word segmentation result into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent;

[0083] Step 303: Input the search field and the data search request into the intent recognition sub-component, and determine the search intent corresponding to the data search request based on the intent recognition sub-component, wherein the intent recognition sub-component is a semantic intent recognition plugin based on NLP.

[0084] By using the pre-processing components in the old search channel, the search field and the data search request are processed to obtain the named font in the search field and the search intent corresponding to the data search request. This makes it easier to perform search optimization on the new search channel in a more scientific way by combining the named font in the search field and the search intent corresponding to the data search request.

[0085] Step 204: Obtain the search recall results of the aforementioned search channels respectively.

[0086] In this embodiment, the step of obtaining the search recall results of the plurality of search channels respectively includes: constructing SQL execution statements adapted to the plurality of search channels respectively based on the search fields and the search intent; performing search processing based on the SQL execution statements adapted to the plurality of search channels respectively and the search functions built into the plurality of search channels respectively, and obtaining the search recall results of the plurality of search channels.

[0087] In this embodiment, after performing the step of obtaining the search recall results of the plurality of search channels respectively, the method further includes: obtaining the distinguishing identifier information of the plurality of search channels; and assigning the distinguishing identifier information to the search recall results corresponding to the plurality of search channels respectively.

[0088] In this embodiment, the step of obtaining the distinguishing identifier information of the plurality of search channels and assigning the distinguishing identifier information to the search recall results corresponding to the plurality of search channels respectively makes it convenient to directly obtain the search recall results corresponding to different search channels based on the distinguishing identifier information, and facilitates the quick acquisition of the search recall results corresponding to the old search channels and the new search channels respectively.

[0089] Step 205: Based on the search recall results, the named entity, the search intent, and the preset search optimization strategy, optimize the new search channel.

[0090] Continue to refer to Figure 4 , Figure 4 yes Figure 2 A flowchart of a specific embodiment of step 205 shown includes:

[0091] Step 401: Using a preset similarity algorithm and the distinguishing identifier information, calculate the similarity between the search recall results of the new search channel and the search recall results of all old search channels.

[0092] Step 402: By cyclic comparison, determine whether the similarity between the search recall results of the current old search channel and the search recall results of the new search channel meets the preset similarity threshold.

[0093] Step 403: If the similarity between the search recall result of the current old search channel and the search recall result of the new search channel meets the similarity threshold, then update the next old search channel to the current old search channel and execute step 402 until all old search channels have been compared with the new search channel, and stop executing step 402.

[0094] Step 404: If the similarity between the search recall results of the current old search channel and the search recall results of the new search channel does not meet the similarity threshold, then the new search channel is optimized based on the search recall results, named entities, and search intent corresponding to the current old search channel.

[0095] By using a cyclical comparison method, the similarity between the search recall results of the new search channel and the search recall results of all old search channels is determined. Based on the comparison results, it is determined whether to optimize the new search channel by combining it with the current old search channels. This facilitates rapid optimization of the new search channel by combining it with old search channels, ensuring that the optimization of the new search channel incorporates the relatively mature old search channels, and indirectly ensuring the accuracy of the search recall results of the optimized new search channel.

[0096] Continue to refer to Figure 5 , Figure 5 yes Figure 4 A flowchart of a specific embodiment of the iterative optimization steps following step 404 is shown, including:

[0097] Step 501: Resend the search field to the new search channel for data search and obtain the latest search recall results corresponding to the new search channel;

[0098] Step 502: Calculate the similarity between the latest search recall result and the search recall results of all old search channels using the similarity algorithm and the distinguishing identifier information.

[0099] Step 503: By comparison, determine whether the similarity between the search recall results of the current old search channel and the latest search recall results meets the similarity threshold.

[0100] Step 504: If the similarity between the search recall result of the current old search channel and the latest search recall result does not meet the similarity threshold, then continue to optimize the search for the new search channel based on the search recall result, named entity and search intent corresponding to the current old search channel.

[0101] Step 505: Repeat steps 501 to 503 until the similarity between the search recall results of all old search channels and the latest search recall results corresponding to the new search channel meets the similarity threshold. Then stop repeating steps 501 to 503 to complete the search optimization for the new search channel.

[0102] An iterative optimization approach is adopted. After each optimization of a new search channel, the search field is resent to the new search channel for data search to obtain the latest search recall results corresponding to the new search channel. The latest search recall results are compared with the search recall results of all old search channels. Based on the comparison results, the new search channel is iteratively optimized until the similarity between the search recall results of all old search channels and the latest search recall results corresponding to the new search channel meets the similarity threshold. At this point, the repeated execution of steps 501 to 503 is stopped, and the search optimization of the new search channel is completed. This ensures that the search recall results of the new search channel are optimized based on all old search channels, making them more scientific and applicable.

[0103] In this embodiment, the step of optimizing the new search channel based on the search recall results, named entities, and search intents corresponding to the current old search channel specifically includes: using transfer learning to migrate the named entities as search entity tags to the new search channel; migrating the search intent to the new search channel and setting the search intent as the search intent of the new search channel; migrating the search recall results to a preset search result cache library in the new search channel, and constructing a knowledge mapping relationship between the named entities, the search intents, and the search recall results within the new search channel.

[0104] By constructing a knowledge mapping relationship between the named entity, the search intent, and the search recall results within the new search channel, it becomes easier to perform a search on the search field again using the new search channel, based directly on the knowledge mapping relationship and using the search intent and the search recall results as deeper search knowledge.

[0105] This application obtains a data search request and a search field sent by a target search terminal; responds to the data search request and sends the search field to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; based on the pre-processing component in the old search channel, it identifies the named entities in the search field and the search intent corresponding to the data search request; it obtains the search recall results of the several search channels respectively; and based on the search recall results, the named entities, the search intent, and a preset search optimization strategy, it optimizes the search for the new search channel. By processing the search field and the data search request through the pre-processing component in the old search channel to obtain the named font in the search field and the search intent corresponding to the data search request, it facilitates subsequent search optimization of the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. Combining the old search channel with the new search channel for search optimization ensures that the search optimization of the new search channel incorporates the relatively mature old search channel, and also indirectly ensures the accuracy of the search recall results of the optimized new search channel.

[0106] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0107] Foundational artificial intelligence technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, large-scale search optimization techniques, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0108] In this embodiment, a data search request and search field sent by a target search terminal are obtained; in response to the data search request, the search field is sent to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; based on the pre-processing component in the old search channel, the named entity in the search field and the search intent corresponding to the data search request are identified; the search recall results of the several search channels are obtained respectively; based on the search recall results, the named entity, the search intent, and a preset search optimization strategy, the new search channel is optimized. By processing the search field and the data search request through the pre-processing component in the old search channel, the named font in the search field and the search intent corresponding to the data search request are obtained, which facilitates subsequent search optimization of the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. Combining the old search channel with the new search channel for search optimization ensures that the search optimization of the new search channel incorporates the relatively mature old search channel, and also indirectly ensures the accuracy of the search recall results of the optimized new search channel.

[0109] Further reference Figure 6 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of a search optimization apparatus, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0110] like Figure 6 As shown, the search optimization device 600 described in this embodiment includes: a search information acquisition module 601, a data search module 602, a preprocessing module 603, a search recall result acquisition module 604, and a new search channel search optimization module 605. Wherein:

[0111] The search information acquisition module 601 is used to acquire the data search request and search fields sent by the target search terminal;

[0112] The data search module 602 is used to respond to the data search request and send the search field to a number of preset search channels for data search respectively, wherein the number of search channels includes old search channels and new search channels;

[0113] The pre-processing module 603 is used to identify the named entities in the search field and the search intent corresponding to the data search request based on the pre-processing components in the old search channel.

[0114] The search recall result acquisition module 604 is used to acquire the search recall results of the plurality of search channels respectively;

[0115] The new search channel search optimization module 605 is used to optimize the new search channel based on the search recall results, the named entity, the search intent, and the preset search optimization strategy.

[0116] In some specific embodiments of this application, the pre-processing module 603 includes: a word segmentation processing submodule, a named entity recognition submodule, and a search intent recognition submodule. Wherein:

[0117] The word segmentation processing submodule is used to input the search field into the word segmentation subcomponent and obtain the word segmentation result output by the word segmentation subcomponent, wherein the word segmentation subcomponent includes the jieba word segmentation plugin;

[0118] The named entity recognition submodule is used to input the word segmentation result into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent;

[0119] The search intent recognition submodule is used to input the search field and the data search request into the intent recognition subcomponent, and determine the search intent corresponding to the data search request based on the intent recognition subcomponent. The intent recognition subcomponent is a semantic intent recognition plugin based on NLP.

[0120] In some specific embodiments of this application, the search optimization device 800 further includes an iterative optimization module, which is used to, after optimizing the new search channel based on the search recall results, named entities, and search intent corresponding to the current old search channel, resend the search field to the new search channel for data search, obtain the latest search recall results corresponding to the new search channel, and calculate the similarity between the latest search recall results and the search recall results of all old search channels using the similarity algorithm and the distinguishing identifier information. Through iterative comparison, it is determined whether the similarity between the search recall results of the current old search channel and the latest search recall results meets the similarity threshold. If the similarity between the search recall results of the current old search channel and the latest search recall results does not meet the similarity threshold, then the search optimization of the new search channel continues based on the search recall results, named entities, and search intent corresponding to the current old search channel until the similarity between the search recall results of all old search channels and the latest search recall results corresponding to the new search channel meets the similarity threshold, at which point the iterative optimization stops, and the search optimization of the new search channel is completed.

[0121] This application obtains a data search request and a search field sent by a target search terminal; responds to the data search request and sends the search field to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; based on the pre-processing component in the old search channel, it identifies the named entities in the search field and the search intent corresponding to the data search request; it obtains the search recall results of the several search channels respectively; and based on the search recall results, the named entities, the search intent, and a preset search optimization strategy, it optimizes the search for the new search channel. By processing the search field and the data search request through the pre-processing component in the old search channel to obtain the named font in the search field and the search intent corresponding to the data search request, it facilitates subsequent search optimization of the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. Combining the old search channel with the new search channel for search optimization ensures that the search optimization of the new search channel incorporates the relatively mature old search channel, and also indirectly ensures the accuracy of the search recall results of the optimized new search channel.

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0123] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0124] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 7 , Figure 7 This is a basic structural block diagram of the computer device in this embodiment.

[0125] The computer device 7 includes a memory 7a, a processor 7b, and a network interface 7c that are interconnected via a system bus. It should be noted that only the computer device 7 with components 7a-7c is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0126] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0127] The memory 7a includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 7a may be an internal storage unit of the computer device 7, such as the hard disk or memory of the computer device 7. In other embodiments, the memory 7a may also be an external storage device of the computer device 7, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 7. Of course, the memory 7a may include both the internal storage unit and its external storage device of the computer device 7. In this embodiment, the memory 7a is typically used to store the operating system and various application software installed on the computer device 7, such as computer-readable instructions for a search optimization method. In addition, the memory 7a can also be used to temporarily store various types of data that have been output or will be output.

[0128] In some embodiments, the processor 7b may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other search optimization chip. The processor 7b is typically used to control the overall operation of the computer device 7. In this embodiment, the processor 7b is used to execute computer-readable instructions stored in the memory 7a or to process data, such as executing computer-readable instructions for the search optimization method.

[0129] The network interface 7c may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 7 and other electronic devices.

[0130] The computer device proposed in this embodiment belongs to the field of financial technology and is applied to the optimization of new search channels in financial business. This application obtains a data search request and search field sent by a target search terminal; responds to the data search request and sends the search field to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; based on the pre-processing component in the old search channel, it identifies the named entities in the search field and the search intent corresponding to the data search request; it obtains the search recall results of the several search channels respectively; and based on the search recall results, the named entities, the search intent, and a preset search optimization strategy, it optimizes the search for the new search channel. By processing the search field and the data search request through the pre-processing component in the old search channel, the named font in the search field and the search intent corresponding to the data search request are obtained, which facilitates subsequent search optimization of the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. By combining existing search channels with new search channels for search optimization, the optimization of new search channels is ensured to be integrated with the relatively mature existing search channels, which also indirectly ensures the accuracy of search recall results after the optimization of new search channels.

[0131] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by a processor to cause the processor to perform the steps of the search optimization method described above.

[0132] The computer-readable storage medium proposed in this embodiment belongs to the field of financial technology and is applied to the optimization of new search channels in financial business. This application obtains a data search request and search field sent by a target search terminal; responds to the data search request and sends the search field to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; based on the pre-processing component in the old search channel, it identifies the named entities in the search field and the search intent corresponding to the data search request; it obtains the search recall results of the several search channels respectively; and based on the search recall results, the named entities, the search intent, and a preset search optimization strategy, it optimizes the search for the new search channel. By processing the search field and the data search request through the pre-processing component in the old search channel, the named font in the search field and the search intent corresponding to the data search request are obtained, facilitating subsequent search optimization of the new search channel by combining the named font in the search field and the search intent corresponding to the data search request. By combining existing search channels with new search channels for search optimization, the optimization of new search channels is ensured to be integrated with the relatively mature existing search channels, which also indirectly ensures the accuracy of search recall results after the optimization of new search channels.

[0133] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0134] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A search optimization method, characterized in that, Includes the following steps: Step 201: Obtain the data search request and search fields sent by the target search terminal; Step 202: In response to the data search request, the search field is sent to several preset search channels for data search, wherein the several search channels include old search channels and new search channels; Step 203: Based on the pre-processing component in the old search channel, identify the named entities in the search field and the search intent corresponding to the data search request; Step 204: Obtain the search recall results of the plurality of search channels respectively. After performing the step of obtaining the search recall results of the plurality of search channels respectively, the method further includes: Obtain the distinguishing identifiers of the aforementioned search channels; The distinguishing identifier information is assigned to the search recall results corresponding to the plurality of search channels respectively; Step 205: Based on the search recall results, the named entities, the search intent, and the preset search optimization strategy, optimize the new search channel. Specifically, Step 401: Using a preset similarity algorithm and the distinguishing identifier information, calculate the similarity between the search recall results of the new search channel and the search recall results of all old search channels. Step 402: By cyclic comparison, determine whether the similarity between the search recall results of the current old search channel and the search recall results of the new search channel meets the preset similarity threshold. Step 403: If the similarity between the search recall result of the current old search channel and the search recall result of the new search channel meets the similarity threshold, then update the next old search channel to the current old search channel and execute step 402 until all old search channels have been compared with the new search channel, and stop executing step 402. Step 404: If the similarity between the search recall results of the current old search channel and the search recall results of the new search channel does not meet the similarity threshold, then the new search channel is optimized based on the search recall results, named entities, and search intent corresponding to the current old search channel. Specifically, the step of optimizing the new search channel based on the search recall results, named entities, and search intent corresponding to the current old search channel is implemented as follows: The named entities are transferred as search entity tags to the new search channel using transfer learning. The search intent is migrated to the new search channel, and the search intent is set as the search intent of the new search channel; The search recall results are migrated to a pre-set search result cache library in the new search channel, and a knowledge mapping relationship is constructed between the named entity, the search intent, and the search recall results within the new search channel.

2. The search optimization method according to claim 1, characterized in that, The pre-processing component includes a word segmentation sub-component, a named entity recognition sub-component, and an intent recognition sub-component. The step of identifying the named entities in the search field and the search intent corresponding to the data search request based on the pre-processing component in the old search channel specifically includes: The search field is input into the word segmentation sub-component to obtain the word segmentation result output by the word segmentation sub-component, wherein the word segmentation sub-component includes the jieba word segmentation plugin; The word segmentation result is input into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent; The search field and the data search request are input into the intent recognition subcomponent. The search intent corresponding to the data search request is determined based on the intent recognition subcomponent. The intent recognition subcomponent is a semantic intent recognition plugin based on NLP.

3. The search optimization method according to claim 1, characterized in that, The step of obtaining the search recall results of the plurality of search channels respectively specifically includes: Based on the search fields and the search intent, construct SQL execution statements adapted to the various search channels respectively; Based on the SQL execution statements adapted to the various search channels and the search functions built into the various search channels, search processing is performed to obtain the search recall results of the various search channels.

4. The search optimization method according to claim 3, characterized in that, After performing the step of optimizing the new search channel based on the search recall results, named entities, and search intent corresponding to the current old search channel, the method further includes: Step 501: Resend the search field to the new search channel for data search and obtain the latest search recall results corresponding to the new search channel; Step 502: Calculate the similarity between the latest search recall result and the search recall results of all old search channels using the similarity algorithm and the distinguishing identifier information; Step 503: By comparison, determine whether the similarity between the search recall results of the current old search channel and the latest search recall results meets the similarity threshold. Step 504: If the similarity between the search recall result of the current old search channel and the latest search recall result does not meet the similarity threshold, then continue to optimize the search for the new search channel based on the search recall result, named entity and search intent corresponding to the current old search channel. Step 505: Repeat steps 501 to 503 until the similarity between the search recall results of all old search channels and the latest search recall results corresponding to the new search channel meets the similarity threshold. Then stop repeating steps 501 to 503 to complete the search optimization for the new search channel.

5. A search optimization device, characterized in that, The search optimization apparatus implements the steps of the search optimization method as described in any one of claims 1 to 4, and the search optimization apparatus includes: The search information acquisition module is used to acquire the data search request and search fields sent by the target search terminal. A data search module is used to respond to the data search request and send the search field to a number of preset search channels for data search, wherein the number of search channels includes old search channels and new search channels; The pre-processing module is used to identify the named entities in the search field and the search intent corresponding to the data search request based on the pre-processing components in the old search channel. The search recall result acquisition module is used to acquire the search recall results of the plurality of search channels respectively; The new search channel search optimization module is used to optimize the new search channel based on the search recall results, the named entities, the search intent, and the preset search optimization strategy.

6. The search optimization apparatus according to claim 5, characterized in that, The aforementioned preprocessing module includes: The word segmentation processing submodule is used to input the search field into the word segmentation subcomponent and obtain the word segmentation result output by the word segmentation subcomponent, wherein the word segmentation subcomponent includes the jieba word segmentation plugin; The named entity recognition submodule is used to input the word segmentation result into the named entity recognition subcomponent to obtain the named font output by the named entity recognition subcomponent; The search intent recognition submodule is used to input the search field and the data search request into the intent recognition subcomponent, and determine the search intent corresponding to the data search request based on the intent recognition subcomponent. The intent recognition subcomponent is a semantic intent recognition plugin based on NLP.

7. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the search optimization method as described in any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the search optimization method as described in any one of claims 1 to 4.

Images