CN117349515A - Search processing method, electronic device and storage medium - Google Patents
Search processing method, electronic device and storage medium Download PDFInfo
- Publication number
- CN117349515A CN117349515A CN202311219863.3A CN202311219863A CN117349515A CN 117349515 A CN117349515 A CN 117349515A CN 202311219863 A CN202311219863 A CN 202311219863A CN 117349515 A CN117349515 A CN 117349515A
- Authority
- CN
- China
- Prior art keywords
- search
- search request
- attribute
- auxiliary information
- commerce service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 62
- 238000004458 analytical method Methods 0.000 claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000000605 extraction Methods 0.000 claims description 64
- 238000002372 labelling Methods 0.000 claims description 63
- 238000012549 training Methods 0.000 claims description 39
- 230000004044 response Effects 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 13
- 238000010801 machine learning Methods 0.000 claims description 5
- 238000000034 method Methods 0.000 abstract description 59
- 230000008569 process Effects 0.000 abstract description 38
- 238000005516 engineering process Methods 0.000 abstract description 17
- 230000006870 function Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000005065 mining Methods 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 6
- 238000003058 natural language processing Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000012896 Statistical algorithm Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a search processing method, electronic equipment and a storage medium, and relates to the fields of large model technology and computer technology. Wherein the method comprises the following steps: acquiring a search request; performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and acquiring a search result corresponding to the search request based on the search request and the auxiliary information. The method and the device solve the technical problem that industry knowledge is difficult to inject into a search processing process in the related technology, so that search accuracy is low.
Description
Technical Field
The present application relates to the field of large model technology and computer technology, and in particular, to a search processing method, an electronic device, and a storage medium.
Background
In recent years, search engine technology based on artificial intelligence models is increasingly applied to various industries. However, conventional search engines are typically implemented based on small models and simple statistical algorithms, which are difficult to adapt to complex scenarios. Currently, although some large model-based search engines (such as differential search indexes, differential Search Index, DSI) exist, existing large model-based search engines often rely entirely on the performance of large models, and it is difficult to combine large models with the features of industry customization and human intervention of traditional search engines, which limits the search accuracy of existing large model-based search engines.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the application provides a search processing method, electronic equipment and a storage medium, which at least solve the technical problem that industry knowledge is difficult to inject into a search processing process in the related technology, so that search accuracy is low.
According to an aspect of an embodiment of the present application, there is provided a search processing method, including: acquiring a search request; performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and acquiring a search result corresponding to the search request based on the search request and the auxiliary information.
According to another aspect of the embodiments of the present application, there is also provided another search processing method, including: acquiring a search request; performing attribute feature analysis on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and performing industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and obtaining a search result corresponding to the search request based on the target document identification.
According to another aspect of the embodiments of the present application, there is further provided a search processing method, including: acquiring an e-commerce service search request; e-business attribute feature analysis is carried out on the E-business service search request, and E-business service auxiliary information associated with the E-business service search request is generated, wherein the E-business service auxiliary information is a target E-business attribute feature combination of E-business service industry knowledge corresponding to the E-business service search request; and acquiring an E-commerce service search result corresponding to the E-commerce service search request based on the E-commerce service search request and the E-commerce service auxiliary information.
According to another aspect of the embodiments of the present application, there is also provided an electronic device, including: a memory storing an executable program; and the processor is used for running a program, wherein the program executes any one of the search processing methods.
According to another aspect of the embodiments of the present application, there is further provided a computer readable storage medium, where the computer readable storage medium includes a stored program, and when the program runs, the device on which the computer readable storage medium is located is controlled to execute any one of the search processing methods described above.
In the embodiment of the application, a search request is firstly acquired, auxiliary information associated with the search request is generated by carrying out attribute feature analysis on the search request, and a search result corresponding to the search request is further acquired based on the search request and the auxiliary information. In the above process, the auxiliary information is at least one target attribute feature of the preset industry knowledge corresponding to the search request, so that the attribute feature corresponding to the preset industry knowledge is introduced into the search process by mining the attribute feature of the search request, the purpose of searching the preset industry knowledge is achieved, the technical effect of improving the accuracy of searching the specific industry knowledge is achieved, and the technical problem that the search accuracy is low due to the fact that the industry knowledge is difficult to inject into the search process in the related technology is solved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a schematic diagram of an application scenario of a search processing method in the present application;
FIG. 2 is a flow chart of a search processing method according to embodiment 1 of the present application;
FIG. 3 is a schematic diagram of an alternative search process according to an embodiment of the present application;
FIG. 4 is a flow chart of another search processing method according to embodiment 2 of the present application;
FIG. 5 is a flow chart of another search processing method according to embodiment 3 of the present application;
fig. 6 is a schematic structural view of a search processing apparatus according to embodiment 4 of the present application;
FIG. 7 is a schematic diagram of an alternative search processing apparatus according to embodiment 4 of the present application;
fig. 8 is a schematic structural view of another search processing apparatus according to embodiment 4 of the present application;
Fig. 9 is a schematic structural view of still another search processing apparatus according to embodiment 4 of the present application;
fig. 10 is a block diagram of a computer terminal according to embodiment 5 of the present application.
Detailed Description
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The technical scheme provided by the application is mainly realized by adopting a large model technology, wherein a large model refers to a deep learning model with large-scale model parameters, and the deep learning model can generally contain hundreds of millions, billions, trillions and even billions of model parameters. The large Model can be called as a Foundation Model, a training Model is performed by using a large-scale unlabeled corpus, a pre-training Model with more than one hundred million parameters is produced, the Model can adapt to a wide downstream task, and the Model has better generalization capability, such as a large-scale language Model (Large Language Model, LLM), a multi-mode pre-training Model and the like.
It should be noted that, when the large model is actually applied, the pretrained model can be finely tuned by a small number of samples, so that the large model can be applied to different tasks. For example, the large model can be widely applied to the fields of natural language processing (Natural Language Processing, NLP), computer vision and the like, and particularly can be applied to the tasks of the fields of computer vision such as visual questions and answers (Visual Question Answering, VQA), image descriptions (ICs), image generation and the like, and can also be widely applied to the tasks of the fields of natural language processing such as emotion classification based on texts, text abstract generation, machine translation and the like. Thus, major application scenarios for large models include, but are not limited to, digital assistants, intelligent robots, searches, online education, office software, electronic commerce, intelligent design, and the like.
First, partial terms or terminology appearing in the course of describing the embodiments of the present application are applicable to the following explanation.
The reasoning process comprises the following steps: the model is subjected to a process of outputting corresponding output through data input in an actual scene.
Large Language Model (LLM): refers to an artificial intelligence model with a degree of general-purpose capabilities.
Differential search index (Differential Search Index, DSI): refers to a search system that uses large models instead of traditional search engines.
Named entity recognition (Named Entity Recognition, NER): refers to a technique for extracting named entities from text.
Text segment (sample): the input initial text, which is a large model, is used to guide the large model to generate specific content.
Feature (feature): refers to attribute information obtained after attribute marking of the query.
Example 1
According to an embodiment of the present application, there is provided a search processing method, it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different from that herein.
Considering that the model parameters of the large model are huge and the computing resources of the mobile terminal are limited, the search processing method provided in the embodiment of the present application can be applied to the application scenario shown in fig. 1, but is not limited thereto. In the application scenario illustrated in fig. 1, the large model is deployed in a server 10, and the server 10 may connect to one or more client devices 20 via a local area network connection, a wide area network connection, an internet connection, or other type of data network, where the client devices 20 may include, but are not limited to: smart phones, tablet computers, notebook computers, palm computers, personal computers, smart home devices, vehicle-mounted devices and the like. The client device 20 can interact with a user through a graphical user interface to realize the invocation of the large model, thereby realizing the method provided by the embodiment of the application.
In the above-described operating environment, the present application provides a search processing method as shown in fig. 2. Fig. 2 is a flowchart of a search processing method according to embodiment 1 of the present application, and as shown in fig. 2, the search processing method includes:
step S21, obtaining a search request;
step S22, carrying out attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
Step S23, based on the search request and the auxiliary information, obtaining a search result corresponding to the search request.
The search request may be a query (query) request in the target application scenario. The above target application scenario may be, but is not limited to: the fields of electronics, education, medical treatment, conferences, social networks, financial products, logistics, navigation, and the like involve scenes based on LLM searching. Correspondingly, the preset industry knowledge is the industry knowledge corresponding to the target application scene.
The auxiliary information is obtained by preset industry knowledge and at least one target attribute feature, wherein the at least one target attribute feature comprises attribute features in multiple dimensions obtained by attribute feature mining of the search request. The combination of the search request and the auxiliary information serves as an input to a search model (which may be LLM) and the search results are obtained from the output of the search model. Therefore, the searching processing process can have good intelligent searching performance in the application scene of the preset industry, namely, can obtain accurate searching results and meets the personalized searching requirements of the specific industry.
In the embodiment of the application, a search request is firstly acquired, auxiliary information associated with the search request is generated by carrying out attribute feature analysis on the search request, and a search result corresponding to the search request is further acquired based on the search request and the auxiliary information. In the above process, the auxiliary information is at least one target attribute feature of the preset industry knowledge corresponding to the search request, so that the attribute feature corresponding to the preset industry knowledge is introduced into the search process by mining the attribute feature of the search request, the purpose of searching the preset industry knowledge is achieved, the technical effect of improving the accuracy of searching the specific industry knowledge is achieved, and the technical problem that the search accuracy is low due to the fact that the industry knowledge is difficult to inject into the search process in the related technology is solved.
In the embodiment of the application, a system formed by the client device and the server may perform the following steps: the method comprises the steps that a client device sends a search request to a server, the server executes a step corresponding to a search processing method, attribute feature analysis is conducted on the search request, auxiliary information related to the search request is generated, the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request, search results corresponding to the search request are obtained based on the search request and the auxiliary information, and the search results are returned to the client (or a use interface for providing a search processing function for the client).
It should be noted that, in the case where the operation resource of the client device can meet the deployment and operation conditions of the large model, the embodiment of the application may be performed in the client device.
In a specific implementation manner of the target application scenario, according to the search processing method provided by the embodiment of the application, industry knowledge corresponding to the target application scenario is injected into the middle reasoning process of the DSI, so that accuracy of a search result is improved, and correlation between the search result and a corresponding industry is improved. In the above specific implementation, the search process may include five stages as shown in fig. 3: an industry knowledge attribute extraction stage, a training small model stage, a feature selection stage, a thinking chain training stage and an online reasoning stage. The search processing method provided in the embodiment of the present application is further described below by taking this specific implementation manner as an example.
In an alternative embodiment, in step S22, attribute feature analysis is performed on the search request, and auxiliary information associated with the search request is generated, including the following method steps:
step S221, industry knowledge attribute extraction is carried out on the search request to obtain an attribute extraction result;
step S222, carrying out attribute labeling on the attribute extraction result to obtain an attribute labeling result, wherein the attribute labeling result is used for indicating knowledge attributes of preset industry knowledge in multiple dimensions;
Step S223, performing feature selection on the attribute labeling result to obtain auxiliary information.
The knowledge attribute corresponding to the attribute labeling result at least comprises: geographic attributes, brand attributes, product attributes, etc. of industry knowledge are preset. In the above alternative embodiment, in the industry knowledge attribute extraction stage in the search process, the industry knowledge attribute is extracted from the search request, and the dimensions corresponding to the industry knowledge attribute are determined by the scene requirement. And in the training small model stage, performing attribute labeling on the extracted multiple dimension industry knowledge attributes to obtain attribute labeling results, wherein the attribute labeling results can contain multiple groups of labeling data, and each group of labeling data comprises the industry knowledge attributes and corresponding labels. In the feature selection stage, feature selection is carried out on the attribute labeling results, the obtained auxiliary information is target labeling data matched with the search request in multiple groups of labeling data, and the target labeling data is considered to be industry knowledge suitable for helping to obtain the search results in the search processing process.
In an alternative embodiment, in step S221, industry knowledge attribute extraction is performed on the search request, and obtaining an attribute extraction result includes at least one of the following steps:
Step S2211, word segmentation processing is carried out on a search request based on preset industry knowledge, and a first attribute extraction result is obtained;
step S2212, carrying out named entity recognition on the search request based on preset industry knowledge to obtain a second attribute extraction result;
step S2213, carrying out synonym expansion on the first attribute extraction result based on preset industry knowledge to obtain a third attribute extraction result;
step S2214, keyword extraction is carried out on the search request based on preset industry knowledge, and a fourth attribute extraction result is obtained;
step S2215, based on the preset industry knowledge, the search request is rewritten in the expression form, and a fifth attribute extraction result is obtained.
As an exemplary embodiment, the following various attributes that help to achieve a link between a search request (e.g., multiple queries) and a target answer text are extracted from preset industry knowledge (e.g., e-commerce knowledge, computer technology knowledge, etc.).
First, word segmentation information. And dividing the search request into a plurality of words according to semantics to obtain word segmentation information. For example, for "on the Yangtze river bridge in Nanjing city" contained in the query sentence, word segmentation information obtained after semantic division is: "in", "Nanjing", "City", "Changjiang bridge", "Shang" ].
Second, named Entity Recognition (NER) information. And identifying industry knowledge entities contained in the search request to obtain NER information. For example, for "apple mobile phone shell" contained in the query sentence, the NER information obtained by performing industry knowledge entity recognition includes: "apple": brand "," mobile phone shell ": product" ].
Third, synonym information. And carrying out synonym expansion processing on the word segmentation information to obtain synonym information corresponding to the search request so as to increase the coverage area and the accuracy of search.
Fourth, keyword (keyword) information. Keywords in a search request (e.g., a query statement) are identified for subsequent use.
Fifth, the information is rewritten. And carrying out expression form rewriting on the search request to obtain rewriting information, wherein the rewriting information comprises a plurality of expression sentences corresponding to the same search request.
Through the steps S221, S2211 to S2215, the embodiment of the application realizes industry knowledge attribute extraction of the search request in multiple dimensions, improves machine understanding of the search request in the search processing process, and further helps to improve accuracy and relevance of the search result.
In an alternative embodiment, in step S222, attribute labeling is performed on the attribute extraction result to obtain an attribute labeling result, including the following method steps:
Step S2221, performing attribute labeling on the attribute extraction result by using a labeling model corresponding to the preset industry knowledge, to obtain an attribute labeling result, where the attribute labeling result includes: a plurality of attribute feature combinations.
As an exemplary embodiment, the labeling model corresponding to the preset industry knowledge may be a pre-trained small model customized by the industry, for example, a word segmentation model, a NER model, etc. of a specified industry. And in a training small model stage in the search processing, performing attribute marking on the attribute extraction result by adopting the pre-training small model to obtain a plurality of attribute feature combinations. For example, a word segmentation model of a specified industry is used to perform attribute marking on a word segmentation result of a search request corresponding to the specified industry, so as to obtain n groups of features { f1, f2, & gt, fn } corresponding to n words in the word segmentation result, where each group of features may include at least one feature, for example, a group fi corresponding to an ith word includes a plurality of features.
In an alternative embodiment, in step S223, feature selection is performed on the attribute labeling result to obtain auxiliary information, including the following method steps:
step S2231, selecting a plurality of candidate feature combinations from attribute labeling results;
Step S2232, calculating information gains of a plurality of candidate feature combinations for a preset result;
step S2233, based on the information gain, selects at least one target attribute feature from the plurality of candidate feature combinations, resulting in the auxiliary information.
As an exemplary embodiment, in a feature selection stage in the search process, for a search request (such as a plurality of queries), a plurality of candidate feature combinations are selected from the corresponding attribute labeling results, for example, m feature combinations are obtained by random sampling. And calculating the information gain of each feature combination to a preset result based on the m feature combinations, wherein the preset result is an expected answer corresponding to the search request. When the search request includes a plurality of queries, the auxiliary information includes at least one target attribute feature F corresponding to each query.
In an alternative embodiment, in step S2232, the information gain of the plurality of candidate feature combinations for the preset result is calculated, including the following method steps:
step S321, obtaining a first probability of obtaining a preset result based on the search request and the auxiliary information prediction, and obtaining a second probability of obtaining the preset result based on the search request prediction;
in step S322, the information gain is calculated by using the first probability and the second probability.
As an exemplary embodiment, when calculating the information gain of m feature combinations to the preset result, the m feature combinations are denoted as F1 to Fm, and for any one of the feature combinations Fi, a first probability P (doc|query, fi) that the preset result is obtained in response to the search request with the auxiliary information and a second probability P (doc|query) that the preset result is obtained in response to the search request without the auxiliary information are calculated. For any one feature combination Fi, the difference between the first probability and the second probability is calculated as the corresponding information gain Hi, i.e., hi=p (doc|query, fi) -P (doc|query).
In an alternative embodiment, at least one target attribute feature is selected from a plurality of candidate feature combinations based on the information gain in step S2233, comprising the method steps of:
step S331, selecting a candidate feature combination with the maximum information gain from a plurality of candidate feature combinations, and obtaining at least one target attribute feature.
As an exemplary embodiment, based on the information gains H1 to Hm corresponding to the m feature combinations, a feature combination corresponding to the maximum value of the information gain is selected as at least one target attribute feature F. For example, with a maxarg () function, the feature combination corresponding to maxarg (Hi) is selected as F.
In an alternative embodiment, in step S23, based on the search request and the auxiliary information, a search result corresponding to the search request is obtained, including the following method steps:
step S231, carrying out industry knowledge reasoning on the search request and the auxiliary information by adopting a differential search index model to obtain a target document identification, wherein the differential search index model is obtained by training a plurality of groups of data through machine learning, the plurality of groups of data are mixed data of first training data and second training data, and the first training data comprise: sample suggestion, sample document identification, second training data includes: sample prompt, at least one target attribute feature corresponding to the sample prompt, and a sample document identification;
step S232, obtaining the search result corresponding to the search request based on the target document identification.
As an exemplary embodiment, the training set for training the Differential Search Index (DSI) model includes first training data (sample, target) and second training data (query, f_target), where f_target is obtained by stitching at least one target attribute feature F corresponding to the query with a sample document identification target.
In the thinking chain training stage in the search process, at least one target attribute feature F is used as a thinking chain to train the DSI model, so that the DSI model can learn the middle reasoning process of industry knowledge, and the search result has more industry relevance.
In an online reasoning stage in search processing, when a user searches through a trained DSI model, firstly, taking a search request (query_user) input by the user as a sample prompt sample, and analyzing the query_user by using an industry knowledge feature combination by adopting the DSI model to obtain at least one target attribute feature F_user corresponding to the query_user; and then, taking a splicing result of the query_user and at least one target attribute feature F_user as a sample prompt, and obtaining a document number corresponding to the query_user through a DSI model, wherein document content corresponding to the document number is taken as a search result corresponding to the query_user. The search query process for the query_user is completed once.
It is easy to notice that, by the above search processing method provided by the embodiment of the present application, the preset industry knowledge is injected into the middle reasoning process of the DSI model, so that the accuracy of the DSI model search result and the relevance with the specified industry are improved.
In an alternative embodiment, a graphical user interface is provided by the terminal device, and the content displayed by the graphical user interface at least partially includes an e-commerce service search dialog, and the search processing method further includes the following method steps:
Step S241, responding to the input operation executed to the E-commerce service search dialog box, determining an E-commerce service search request;
step S242, in response to the sending operation performed on the e-commerce service search dialog, performing e-commerce service attribute feature analysis on the e-commerce service search request to generate e-commerce service auxiliary information, and obtaining an e-commerce service search result based on the e-commerce service search request and the e-commerce service auxiliary information;
step S243, displaying the E-commerce service search result in the E-commerce service search dialog.
In the above optional embodiment, the e-commerce service search dialog box may be used to implement a dedicated e-commerce service search function corresponding to a preset industry, for example, an intelligent question-answering function set in an application scenario of the preset industry, where the intelligent question-answering function is not limited to text question-answering, and the above embodiment may be combined with technologies such as voice-to-text, text-to-voice, video-to-text, text-to-video, and so on to implement a voice question-answering function, a video question-answering function, or a virtual reality/augmented reality question-answering function.
When an input operation is detected that acts on an e-commerce services search dialog, an e-commerce services search request is determined from input content within the e-commerce services search dialog, which may include at least one query statement.
When detecting the sending operation acting on the e-commerce service search dialog box, triggering a search event, performing e-commerce service attribute feature analysis on the e-commerce service search request to generate e-commerce service auxiliary information, acquiring an e-commerce service search result based on the e-commerce service search request and the e-commerce service auxiliary information, and automatically triggering an event for displaying the e-commerce service search result in the e-commerce service search dialog box. The form of presenting the e-commerce service search results to the user may be: is displayed in a graphical user interface in text form or in picture form, converted into audio to be output through an audio output device, and converted into video to be displayed in the graphical user interface.
The input operation and the transmission operation may be operations in which a user touches the display screen of the terminal device with a finger and touches the terminal device. The touch operation may include single-point touch, multi-point touch, where the touch operation of each touch point may include clicking, long pressing, heavy pressing, swiping, and the like. The input operation and the transmission operation may be touch operation performed by an input device such as a mouse or a keyboard.
Through the steps S241 to S243, the search processing method provided in the embodiment of the present application may be run on the client, and a visual interaction manner is embodied to implement a search processing function, where the visual interaction manner may provide a question-answer scenario of a preset industry to a user through the client more intuitively, so that the user may more conveniently provide an e-commerce service search request and trigger a query of an e-commerce service search result.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and provide corresponding operation entries for the user to select authorization or rejection.
It should be noted that, for simplicity of description, the foregoing method embodiments are all expressed as a series of action combinations, but it should be understood by those skilled in the art that the present application is not limited by the order of actions described, as some steps may be performed in other order or simultaneously in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required in the present application.
From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus a necessary general hardware platform, but that it may also be implemented by means of hardware. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method described in the embodiments of the present application.
Example 2
In the operating environment as in example 1, the present application provides another search processing method as shown in fig. 4. Fig. 4 is a flowchart of another search processing method according to embodiment 2 of the present application, as shown in fig. 4, including:
step S41, obtaining a search request;
step S42, performing attribute feature analysis on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and performing industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
step S43, obtaining the search result corresponding to the search request based on the target document identification.
The search request may be a query (query) request in the target application scenario. The above target application scenario may be, but is not limited to: the fields of electronics, education, medical treatment, conferences, social networks, financial products, logistics, navigation, and the like involve scenes based on LLM searching. Correspondingly, the preset industry knowledge is the industry knowledge corresponding to the target application scene.
The auxiliary information is obtained by preset industry knowledge and at least one target attribute feature, wherein the target attribute feature comprises a plurality of attribute features obtained by attribute feature mining of the search request. The combination of the search request and the auxiliary information is used as input of a Differential Search Index (DSI) model, the DSI model conducts industry knowledge reasoning on the search request and the auxiliary information to output a target document identification, and the target document identification and corresponding document content are used as search results corresponding to the search request.
Therefore, the searching processing process can have good intelligent searching performance in the application scene of the preset industry, namely, can obtain accurate searching results and meets the personalized searching requirements of the specific industry.
In the embodiment of the application, firstly, a search request is acquired, attribute feature analysis is performed on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and industry knowledge reasoning is performed on the search request and the auxiliary information to output a target document identification, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request, and a search result corresponding to the search request is further acquired based on the target document identification. In the above process, the auxiliary information is at least one target attribute feature of the preset industry knowledge corresponding to the search request, so that the attribute feature corresponding to the preset industry knowledge is introduced into the search process by mining the attribute feature of the search request, the purpose of searching the preset industry knowledge is achieved, the technical effect of improving the accuracy of searching the specific industry knowledge is achieved, and the technical problem that the search accuracy is low due to the fact that the industry knowledge is difficult to inject into the search process in the related technology is solved.
In the embodiment of the application, a system formed by the client device and the server may perform the following steps: the method comprises the steps that a client device sends a search request to a server, the server executes a step corresponding to a search processing method, a differential search index model is adopted to conduct attribute feature analysis on the search request to generate auxiliary information related to the search request, industry knowledge reasoning is conducted on the search request and the auxiliary information to output target document identification, the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request, search results corresponding to the search request are obtained based on the target document identification, and the search results are returned to the client (or a using interface for providing a search processing function for the client).
It should be noted that, in the case where the operation resource of the client device can meet the deployment and operation conditions of the large model, the embodiment of the application may be performed in the client device.
In an alternative embodiment, in step S42, the industry knowledge reasoning is performed on the search request and the auxiliary information by using the differential search index model to output the target document identification, including the following method steps:
Step S421, determining initial prompt contents through a search request, wherein the initial prompt contents are configured based on a preset prompt template;
step S422, the initial prompt content and the auxiliary information are spliced to obtain target prompt content;
step S423, carrying out industry knowledge reasoning on the target prompt template by adopting the differential search index model to obtain a target document identification.
In an application scene, determining initial prompt content corresponding to a search request (query) (the initial prompt content can be text segment prompt), splicing the initial prompt content with the auxiliary information to obtain target prompt content, wherein the auxiliary information is obtained by at least one target attribute feature (marked as F) of preset industry knowledge corresponding to the search request, the target prompt content can be expressed as query_F, and performing industry knowledge reasoning on the target prompt content query_F based on the preset industry knowledge by adopting a DSI model with completed training to obtain a target document identifier so as to obtain a search result corresponding to the search request (query).
It is easy to notice that, by the above search processing method provided by the embodiment of the present application, the preset industry knowledge is injected into the middle reasoning process of the DSI model, so that the accuracy of the DSI model search result and the relevance with the specified industry are improved.
It should be noted that, the preferred implementation manner of this embodiment may be referred to the related description in embodiment 1, and will not be repeated here.
Example 3
In the operating environment as in example 1, the present application provides another search processing method as shown in fig. 5. Fig. 5 is a flowchart of another search processing method according to embodiment 3 of the present application, as shown in fig. 5, the search processing method including:
step S51, acquiring an E-commerce service search request;
step S52, carrying out E-commerce attribute feature analysis on the E-commerce service search request to generate E-commerce service auxiliary information associated with the E-commerce service search request, wherein the E-commerce service auxiliary information is at least one target E-commerce attribute feature of E-commerce service industry knowledge corresponding to the E-commerce service search request;
step S53, based on the E-commerce service search request and the E-commerce service auxiliary information, acquiring an E-commerce service search result corresponding to the E-commerce service search request.
The e-commerce service search request may be a query (query) request in an e-commerce service application scenario. Correspondingly, the e-commerce service industry knowledge is industry knowledge corresponding to the e-commerce service application scene, for example, dialogue question-answer knowledge, user comment knowledge and the like.
The e-commerce service auxiliary information is obtained by e-commerce service industry knowledge and at least one target e-commerce attribute feature, wherein the target attribute feature comprises a plurality of attribute features obtained by attribute feature mining of an e-commerce service search request. The combination of the e-commerce service search request and the e-commerce service auxiliary information is used as input of a search model (which can be LLM), and the e-commerce service search result is obtained by output of the search model. Therefore, the search processing process can have good intelligent search performance in an application scene of a preset industry, namely, can obtain accurate E-commerce service search results and meets the personalized search requirements of the E-commerce service industry.
In the embodiment of the application, an e-commerce service search request is firstly acquired, e-commerce attribute feature analysis is performed on the e-commerce service search request, e-commerce service auxiliary information associated with the e-commerce service search request is generated, and an e-commerce service search result corresponding to the e-commerce service search request is further acquired based on the e-commerce service search request and the e-commerce service auxiliary information. In the above process, the e-commerce service auxiliary information is at least one target e-commerce attribute feature of the e-commerce service industry knowledge corresponding to the e-commerce service search request, so that the search processing method provided by the application introduces the attribute feature corresponding to the e-commerce service industry knowledge into the search processing by mining the attribute feature of the e-commerce service search request, thereby achieving the purpose of searching the e-commerce service industry knowledge, further achieving the technical effect of improving the accuracy of searching the e-commerce service industry knowledge, and further solving the technical problem that the search accuracy is lower due to the fact that the industry knowledge is difficult to inject into the search processing process in the related technology.
In the embodiment of the application, a system formed by the client device and the server may perform the following steps: the client device executes the steps corresponding to the search processing method for sending the E-commerce service search request to the server, the server executes the steps corresponding to the search processing method for carrying out E-commerce attribute feature analysis on the E-commerce service search request to generate E-commerce service auxiliary information related to the E-commerce service search request, wherein the E-commerce service auxiliary information is at least one target E-commerce attribute feature of E-commerce service industry knowledge corresponding to the E-commerce service search request, the E-commerce service search result corresponding to the E-commerce service search request is obtained based on the E-commerce service search request and the E-commerce service auxiliary information, and the E-commerce service search result is returned to the client (or a use interface for providing a search processing function for the client).
It should be noted that, in the case where the operation resource of the client device can meet the deployment and operation conditions of the large model, the embodiment of the application may be performed in the client device.
In addition, the search processing method provided in embodiment 1 may be extended to legal question-answering scenes, medical service scenes, educational question-answering scenes, and the like, in addition to the e-commerce service scenes.
In a legal question-answering scenario, the search processing method comprises the following steps: acquiring legal question-answer search requests; e-business attribute feature analysis is carried out on the legal question-answer search request, and legal question-answer auxiliary information associated with the legal question-answer search request is generated, wherein the legal question-answer auxiliary information is at least one target E-business attribute feature of legal question-answer industry knowledge corresponding to the legal question-answer search request; and acquiring legal question-answer search results corresponding to the legal question-answer search request based on the legal question-answer search request and the legal question-answer auxiliary information.
In a medical service scenario, a search processing method includes: acquiring a medical service search request; e-business attribute feature analysis is carried out on the medical service search request, and medical service auxiliary information associated with the medical service search request is generated, wherein the medical service auxiliary information is at least one target E-business attribute feature of medical service industry knowledge corresponding to the medical service search request; and acquiring medical service search results corresponding to the medical service search request based on the medical service search request and the medical service auxiliary information.
In an educational question-answer scenario, the search processing method comprises the following steps: acquiring an educational question-answer search request; e-business attribute feature analysis is carried out on the educational question and answer search request, educational question and answer auxiliary information associated with the educational question and answer search request is generated, wherein the educational question and answer auxiliary information is at least one target E-business attribute feature of educational question and answer industry knowledge corresponding to the educational question and answer search request; based on the educational question and answer search request and the educational question and answer auxiliary information, educational question and answer search results corresponding to the educational question and answer search request are obtained.
It should be noted that, the preferred implementation manner of this embodiment may be referred to the related description in embodiment 1, and will not be repeated here.
Example 4
According to the embodiment of the application, an embodiment of a device for implementing the search processing method is also provided. Fig. 6 is a schematic structural diagram of a search processing apparatus according to embodiment 4 of the present application, as shown in fig. 6, the apparatus includes:
an obtaining module 601, configured to obtain a search request;
the generating module 602 is configured to perform attribute feature analysis on the search request, and generate auxiliary information associated with the search request, where the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
the processing module 603 is configured to obtain a search result corresponding to the search request based on the search request and the auxiliary information.
Optionally, the generating module 602 is further configured to: industry knowledge attribute extraction is carried out on the search request, and an attribute extraction result is obtained; performing attribute labeling on the attribute extraction result to obtain an attribute labeling result, wherein the attribute labeling result is used for indicating knowledge attributes of preset industry knowledge in multiple dimensions; and performing feature selection on the attribute labeling result to obtain auxiliary information.
Optionally, the generating module 602 is further configured to: word segmentation processing is carried out on the search request based on preset industry knowledge, and a first attribute extraction result is obtained; carrying out named entity recognition on the search request based on preset industry knowledge to obtain a second attribute extraction result; carrying out synonym expansion on the first attribute extraction result based on preset industry knowledge to obtain a third attribute extraction result; keyword extraction is carried out on the search request based on preset industry knowledge, and a fourth attribute extraction result is obtained; and based on preset industry knowledge, carrying out expression form rewriting on the search request to obtain a fifth attribute extraction result.
Optionally, the generating module 602 is further configured to: and carrying out attribute labeling on the attribute extraction result by adopting a labeling model corresponding to preset industry knowledge to obtain an attribute labeling result, wherein the attribute labeling result comprises the following steps: a plurality of attribute feature combinations.
Optionally, the generating module 602 is further configured to: selecting a plurality of candidate feature combinations from the attribute labeling results; calculating information gains of a plurality of candidate feature combinations for a preset result; and selecting at least one target attribute feature from the plurality of candidate feature combinations based on the information gain to obtain auxiliary information.
Optionally, the generating module 602 is further configured to: acquiring a first probability of obtaining a preset result based on search request and auxiliary information prediction, and acquiring a second probability of obtaining the preset result based on search request prediction; and calculating the information gain by using the first probability and the second probability.
Optionally, the generating module 602 is further configured to: and selecting a candidate feature combination with the maximum information gain from the plurality of candidate feature combinations to obtain at least one target attribute feature.
Optionally, the processing module 603 is further configured to: and carrying out industry knowledge reasoning on the search request and the auxiliary information by adopting a differential search index model to obtain a target document identification, wherein the differential search index model is obtained by utilizing a plurality of groups of data through machine learning training, the plurality of groups of data are mixed data of first training data and second training data, and the first training data comprise: sample suggestion, sample document identification, second training data includes: sample prompt, at least one target attribute feature corresponding to the sample prompt, and a sample document identification; and obtaining a search result corresponding to the search request based on the target document identification.
Alternatively, fig. 7 is a schematic structural diagram of an alternative search processing apparatus according to embodiment 4 of the present application, and as shown in fig. 7, the apparatus includes, in addition to all the modules shown in fig. 6: the visualization module 604. Providing, by the terminal device, a graphical user interface, wherein the content displayed by the graphical user interface at least partially includes an e-commerce service search dialog, and the visualization module 604 is configured to: determining a search request in response to an input operation performed in response to the e-commerce service search dialog; performing attribute feature analysis on the search request to generate auxiliary information in response to a transmission operation performed on the e-commerce service search dialog, and acquiring a search result based on the search request and the auxiliary information; the search results are displayed within an e-commerce services search dialog.
In the embodiment of the application, firstly, a search request is acquired through an acquisition module, attribute feature analysis is performed on the search request through a generation module, auxiliary information related to the search request is generated, and further, a processing module is adopted to acquire a search result corresponding to the search request based on the search request and the auxiliary information. In the above process, the auxiliary information is at least one target attribute feature of the preset industry knowledge corresponding to the search request, so that the attribute feature corresponding to the preset industry knowledge is introduced into the search process by mining the attribute feature of the search request, the purpose of searching the preset industry knowledge is achieved, the technical effect of improving the accuracy of searching the specific industry knowledge is achieved, and the technical problem that the search accuracy is low due to the fact that the industry knowledge is difficult to inject into the search process in the related technology is solved.
Here, the acquiring module 601, the generating module 602, and the processing module 603 correspond to steps S21 to S23 in embodiment 1, and the three modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to those disclosed in embodiment 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (for example, the memory 104) and processed by one or more processors (for example, the processors 102a,102b, … …,102 n), and the above-mentioned modules may also be executed as a part of the apparatus in the computer terminal 10 provided in embodiment 1.
According to an embodiment of the present application, another embodiment of an apparatus for implementing the search processing method in the foregoing embodiment 2 is also provided. Fig. 8 is a schematic structural diagram of another search processing apparatus according to embodiment 4 of the present application, as shown in fig. 8, including:
an obtaining module 801, configured to obtain a search request;
the generating module 802 is configured to perform attribute feature analysis on a search request by using a differential search index model to generate auxiliary information associated with the search request, and perform industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, where the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
and the processing module 803 is used for acquiring the search result corresponding to the search request based on the target document identification.
Optionally, the generating module 802 is further configured to: determining initial prompt contents through a search request, wherein the initial prompt contents are configured based on a preset prompt template; splicing the initial prompt content and the auxiliary information to obtain a target prompt template; and carrying out industry knowledge reasoning on the target prompt template by adopting the differential search index model to obtain a target document identification.
Here, the above-mentioned obtaining module 801, generating module 802, and processing module 803 correspond to step S41 to step S43 in embodiment 2, and the three modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to those disclosed in embodiment 2 above. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (for example, the memory 104) and processed by one or more processors (for example, the processors 102a,102b, … …,102 n), and the above-mentioned modules may also be executed as a part of the apparatus in the computer terminal 10 provided in embodiment 1.
According to an embodiment of the present application, another embodiment of an apparatus for implementing the search processing method in the foregoing embodiment 3 is also provided. Fig. 9 is a schematic structural diagram of still another search processing apparatus according to embodiment 4 of the present application, as shown in fig. 9, including:
an acquiring module 901, configured to acquire an e-commerce service search request;
the generating module 902 is configured to perform an e-commerce attribute feature analysis on the e-commerce service search request, and generate e-commerce service auxiliary information associated with the e-commerce service search request, where the e-commerce service auxiliary information is at least one target e-commerce attribute feature of e-commerce service industry knowledge corresponding to the e-commerce service search request;
The obtaining module 903 is configured to obtain an e-commerce service search result corresponding to the e-commerce service search request based on the e-commerce service search request and the e-commerce service auxiliary information.
Here, the above-mentioned obtaining module 901, generating module 902 and processing module 903 correspond to steps S51 to S53 in embodiment 3, and the three modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to those disclosed in embodiment 3 above. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (for example, the memory 104) and processed by one or more processors (for example, the processors 102a,102b, … …,102 n), and the above-mentioned modules may also be executed as a part of the apparatus in the computer terminal 10 provided in embodiment 1.
It should be noted that, the preferred implementation manner of this embodiment may be referred to the related description in embodiment 1, embodiment 2 or embodiment 3, and will not be described herein.
Example 5
According to the embodiment of the application, there is further provided a computer terminal, which may be any one of the computer terminal devices in the computer terminal group. Alternatively, in the present embodiment, the above-described computer terminal may be replaced with a terminal device such as a mobile terminal.
Alternatively, in this embodiment, the above-mentioned computer terminal may be located in at least one network device among a plurality of network devices of the computer network.
In this embodiment, the above-mentioned computer terminal may execute the program code of the following steps in the search processing method: acquiring a search request; performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and acquiring a search result corresponding to the search request based on the search request and the auxiliary information.
Alternatively, fig. 10 is a block diagram of a computer terminal according to embodiment 5 of the present application, and as shown in fig. 10, the computer terminal 100 may include: one or more (only one is shown) processors 1002, memory 1004, a memory controller 1006, and a peripheral interface 1008, where the peripheral interface 1008 interfaces with a radio frequency module, an audio module, and a display.
The memory 1004 may be used to store software programs and modules, such as program instructions/modules corresponding to the search processing method and apparatus in the embodiments of the present application, and the processor executes the software programs and modules stored in the memory, thereby performing various functional applications and data processing, that is, implementing the search processing method described above. Memory 1004 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 1004 may further include memory located remotely from the processor, which may be connected to the computer terminal 100 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The processor 1002 may call the information stored in the memory and the application program through the transmission device to perform the following steps: acquiring a search request; performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and acquiring a search result corresponding to the search request based on the search request and the auxiliary information.
Optionally, the processor 1002 may further execute program code for: industry knowledge attribute extraction is carried out on the search request, and an attribute extraction result is obtained; performing attribute labeling on the attribute extraction result to obtain an attribute labeling result, wherein the attribute labeling result is used for indicating knowledge attributes of preset industry knowledge in multiple dimensions; and performing feature selection on the attribute labeling result to obtain auxiliary information.
Optionally, the processor 1002 may further execute program code for: word segmentation processing is carried out on the search request based on preset industry knowledge, and a first attribute extraction result is obtained; carrying out named entity recognition on the search request based on preset industry knowledge to obtain a second attribute extraction result; carrying out synonym expansion on the first attribute extraction result based on preset industry knowledge to obtain a third attribute extraction result; keyword extraction is carried out on the search request based on preset industry knowledge, and a fourth attribute extraction result is obtained; and based on preset industry knowledge, carrying out expression form rewriting on the search request to obtain a fifth attribute extraction result.
Optionally, the processor 1002 may further execute program code for: and carrying out attribute labeling on the attribute extraction result by adopting a labeling model corresponding to preset industry knowledge to obtain an attribute labeling result, wherein the attribute labeling result comprises the following steps: a plurality of attribute feature combinations.
Optionally, the processor 1002 may further execute program code for: selecting a plurality of candidate feature combinations from the attribute labeling results; calculating information gains of a plurality of candidate feature combinations for a preset result; and selecting at least one target attribute feature from the plurality of candidate feature combinations based on the information gain to obtain auxiliary information.
Optionally, the processor 1002 may further execute program code for: acquiring a first probability of obtaining a preset result based on search request and auxiliary information prediction, and acquiring a second probability of obtaining the preset result based on search request prediction; and calculating the information gain by using the first probability and the second probability.
Optionally, the processor 1002 may further execute program code for: and selecting a candidate feature combination with the maximum information gain from the plurality of candidate feature combinations to obtain at least one target attribute feature.
Optionally, the processor 1002 may further execute program code for: and carrying out industry knowledge reasoning on the search request and the auxiliary information by adopting a differential search index model to obtain a target document identification, wherein the differential search index model is obtained by utilizing a plurality of groups of data through machine learning training, the plurality of groups of data are mixed data of first training data and second training data, and the first training data comprise: sample suggestion, sample document identification, second training data includes: sample prompt, at least one target attribute feature corresponding to the sample prompt, and a sample document identification; and obtaining a search result corresponding to the search request based on the target document identification.
Optionally, the processor 1002 may further execute program code for: determining a search request in response to an input operation performed in response to the e-commerce service search dialog; performing attribute feature analysis on the search request to generate auxiliary information in response to a transmission operation performed on the e-commerce service search dialog, and acquiring a search result based on the search request and the auxiliary information; the search results are displayed within an e-commerce services search dialog.
The processor 1002 may call the information stored in the memory and the application program through the transmission device to perform the following steps: acquiring a search request; performing attribute feature analysis on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and performing industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and obtaining a search result corresponding to the search request based on the target document identification.
Optionally, the processor 1002 may further execute program code for: determining initial prompt contents through a search request, wherein the initial prompt contents are configured based on a preset prompt template; splicing the initial prompt content and the auxiliary information to obtain a target prompt template; and carrying out industry knowledge reasoning on the target prompt template by adopting the differential search index model to obtain a target document identification.
The processor 1002 may call the information stored in the memory and the application program through the transmission device to perform the following steps: acquiring an e-commerce service search request; e-business attribute feature analysis is carried out on the E-business service search request, and E-business service auxiliary information associated with the E-business service search request is generated, wherein the E-business service auxiliary information is at least one target E-business attribute feature of E-business service industry knowledge corresponding to the E-business service search request; and acquiring an E-commerce service search result corresponding to the E-commerce service search request based on the E-commerce service search request and the E-commerce service auxiliary information.
By adopting the embodiment of the application, a scheme of a computer terminal for realizing a search processing method is provided. Firstly, acquiring a search request, generating auxiliary information associated with the search request by carrying out attribute feature analysis on the search request, and further acquiring a search result corresponding to the search request based on the search request and the auxiliary information. In the above process, the auxiliary information is at least one target attribute feature of the preset industry knowledge corresponding to the search request, so that the attribute feature corresponding to the preset industry knowledge is introduced into the search process by mining the attribute feature of the search request, the purpose of searching the preset industry knowledge is achieved, the technical effect of improving the accuracy of searching the specific industry knowledge is achieved, and the technical problem that the search accuracy is low due to the fact that the industry knowledge is difficult to inject into the search process in the related technology is solved.
It will be appreciated by those skilled in the art that the configuration shown in fig. 10 is only illustrative, and the computer terminal may be a terminal device such as a smart phone (e.g. an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, and a mobile internet device (Mobile Internet Devices, MID). Fig. 10 does not limit the structure of the computer terminal. For example, the computer terminal 100 may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in fig. 10, or have a different configuration than shown in fig. 10.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, ROM, RAM, magnetic or optical disk, etc.
Example 6
According to an embodiment of the present application, there is also provided a computer-readable storage medium. Alternatively, in this embodiment, the above-described storage medium may be used to store the program code executed by the search processing method provided in embodiment 1, embodiment 2, or embodiment 3 described above.
Alternatively, in this embodiment, the storage medium may be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: acquiring a search request; performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and acquiring a search result corresponding to the search request based on the search request and the auxiliary information.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: industry knowledge attribute extraction is carried out on the search request, and an attribute extraction result is obtained; performing attribute labeling on the attribute extraction result to obtain an attribute labeling result, wherein the attribute labeling result is used for indicating knowledge attributes of preset industry knowledge in multiple dimensions; and performing feature selection on the attribute labeling result to obtain auxiliary information.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: word segmentation processing is carried out on the search request based on preset industry knowledge, and a first attribute extraction result is obtained; carrying out named entity recognition on the search request based on preset industry knowledge to obtain a second attribute extraction result; carrying out synonym expansion on the first attribute extraction result based on preset industry knowledge to obtain a third attribute extraction result; keyword extraction is carried out on the search request based on preset industry knowledge, and a fourth attribute extraction result is obtained; and based on preset industry knowledge, carrying out expression form rewriting on the search request to obtain a fifth attribute extraction result.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: and carrying out attribute labeling on the attribute extraction result by adopting a labeling model corresponding to preset industry knowledge to obtain an attribute labeling result, wherein the attribute labeling result comprises the following steps: a plurality of attribute feature combinations.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: selecting a plurality of candidate feature combinations from the attribute labeling results; calculating information gains of a plurality of candidate feature combinations for a preset result; and selecting at least one target attribute feature from the plurality of candidate feature combinations based on the information gain to obtain auxiliary information.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: acquiring a first probability of obtaining a preset result based on search request and auxiliary information prediction, and acquiring a second probability of obtaining the preset result based on search request prediction; and calculating the information gain by using the first probability and the second probability.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: and selecting a candidate feature combination with the maximum information gain from the plurality of candidate feature combinations to obtain at least one target attribute feature.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: and carrying out industry knowledge reasoning on the search request and the auxiliary information by adopting a differential search index model to obtain a target document identification, wherein the differential search index model is obtained by utilizing a plurality of groups of data through machine learning training, the plurality of groups of data are mixed data of first training data and second training data, and the first training data comprise: sample suggestion, sample document identification, second training data includes: sample prompt, at least one target attribute feature corresponding to the sample prompt, and a sample document identification; and obtaining a search result corresponding to the search request based on the target document identification.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: determining a search request in response to an input operation performed in response to the e-commerce service search dialog; performing attribute feature analysis on the search request to generate auxiliary information in response to a transmission operation performed on the e-commerce service search dialog, and acquiring a search result based on the search request and the auxiliary information; the search results are displayed within an e-commerce services search dialog.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: acquiring a search request; performing attribute feature analysis on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and performing industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request; and obtaining a search result corresponding to the search request based on the target document identification.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: determining initial prompt contents through a search request, wherein the initial prompt contents are configured based on a preset prompt template; splicing the initial prompt content and the auxiliary information to obtain a target prompt template; and carrying out industry knowledge reasoning on the target prompt template by adopting the differential search index model to obtain a target document identification.
Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the steps of: acquiring an e-commerce service search request; e-business attribute feature analysis is carried out on the E-business service search request, and E-business service auxiliary information associated with the E-business service search request is generated, wherein the E-business service auxiliary information is at least one target E-business attribute feature of E-business service industry knowledge corresponding to the E-business service search request; and acquiring an E-commerce service search result corresponding to the E-commerce service search request based on the E-commerce service search request and the E-commerce service auxiliary information.
The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, ROM, RAM, a mobile hard disk, a magnetic disk or an optical disk.
The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application and are intended to be comprehended within the scope of the present application.
Claims (14)
1. A search processing method, comprising:
acquiring a search request;
performing attribute feature analysis on the search request to generate auxiliary information associated with the search request, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
and acquiring a search result corresponding to the search request based on the search request and the auxiliary information.
2. The search processing method according to claim 1, wherein performing attribute feature analysis on the search request, generating the auxiliary information associated with the search request includes:
carrying out industry knowledge attribute extraction on the search request to obtain an attribute extraction result;
performing attribute labeling on the attribute extraction result to obtain an attribute labeling result, wherein the attribute labeling result is used for indicating knowledge attributes of the preset industry knowledge in multiple dimensions;
And performing feature selection on the attribute labeling result to obtain the auxiliary information.
3. The search processing method according to claim 2, wherein the search request is subjected to industry knowledge attribute extraction, and the attribute extraction result includes one or more of the following:
based on the preset industry knowledge, word segmentation processing is carried out on the search request to obtain a first attribute extraction result;
carrying out named entity recognition on the search request based on the preset industry knowledge to obtain a second attribute extraction result;
carrying out synonym expansion on the first attribute extraction result based on the preset industry knowledge to obtain a third attribute extraction result;
based on the preset industry knowledge, extracting keywords from the search request to obtain a fourth attribute extraction result;
and based on the preset industry knowledge, carrying out expression form rewriting on the search request to obtain a fifth attribute extraction result.
4. The search processing method according to claim 2, wherein performing attribute labeling on the attribute extraction result to obtain the attribute labeling result includes:
and carrying out attribute labeling on the attribute extraction result by adopting a labeling model corresponding to the preset industry knowledge to obtain the attribute labeling result, wherein the attribute labeling result comprises: a plurality of attribute feature combinations.
5. The search processing method according to claim 2, wherein performing feature selection on the attribute labeling result to obtain the auxiliary information includes:
selecting a plurality of candidate feature combinations from the attribute labeling results;
calculating information gains of the candidate feature combinations for a preset result;
and selecting the at least one target attribute feature from the plurality of candidate feature combinations based on the information gain to obtain the auxiliary information.
6. The search processing method of claim 5, wherein calculating the information gain of the plurality of candidate feature combinations for the preset result comprises:
acquiring a first probability of obtaining the preset result based on the search request and the auxiliary information prediction, and acquiring a second probability of obtaining the preset result based on the search request prediction;
and calculating the information gain by using the first probability and the second probability.
7. The search processing method of claim 5, wherein selecting the at least one target attribute feature from the plurality of candidate feature combinations based on the information gain comprises:
And selecting the candidate feature combination with the maximum information gain from the plurality of candidate feature combinations to obtain the at least one target attribute feature.
8. The search processing method according to claim 1, wherein acquiring search results corresponding to the search request based on the search request and the auxiliary information comprises:
and carrying out industry knowledge reasoning on the search request and the auxiliary information by adopting a differential search index model to obtain a target document identification, wherein the differential search index model is obtained by utilizing a plurality of groups of data through machine learning training, the plurality of groups of data are mixed data of first training data and second training data, and the first training data comprise: sample suggestion, sample document identification, second training data includes: the sample prompt, at least one target attribute feature corresponding to the sample prompt, and the sample document identification;
and acquiring the search result corresponding to the search request based on the target document identification.
9. The search processing method of claim 1, wherein a graphical user interface is provided by the terminal device, the graphical user interface displaying content at least partially comprising an e-commerce service search dialog, the search processing method further comprising:
Determining an e-commerce service search request in response to an input operation performed by the e-commerce service search dialog;
responding to the sending operation executed by the e-commerce service searching dialog box, performing e-commerce service attribute feature analysis on the e-commerce service searching request to generate the e-commerce service auxiliary information, and acquiring the e-commerce service searching result based on the e-commerce service searching request and the e-commerce service auxiliary information;
and displaying the E-commerce service search results in the E-commerce service search dialog box.
10. A search processing method, comprising:
acquiring a search request;
performing attribute feature analysis on the search request by adopting a differential search index model to generate auxiliary information related to the search request, and performing industry knowledge reasoning on the search request and the auxiliary information to output a target document identifier, wherein the auxiliary information is at least one target attribute feature of preset industry knowledge corresponding to the search request;
and acquiring a search result corresponding to the search request based on the target document identification.
11. The search processing method of claim 10, wherein employing the differential search index model to conduct industry knowledge reasoning on the search request and the auxiliary information to output the target document identification comprises:
Determining initial prompt content through the search request, wherein the initial prompt content is configured based on a preset prompt template;
splicing the initial prompt content and the auxiliary information to obtain target prompt content;
and carrying out industry knowledge reasoning on the target prompt template by adopting the differential search index model to obtain the target document identification.
12. A search processing method, comprising:
acquiring an e-commerce service search request;
e-commerce attribute feature analysis is carried out on the E-commerce service search request, and E-commerce service auxiliary information associated with the E-commerce service search request is generated, wherein the E-commerce service auxiliary information is at least one target E-commerce attribute feature of E-commerce service industry knowledge corresponding to the E-commerce service search request;
and acquiring an E-commerce service search result corresponding to the E-commerce service search request based on the E-commerce service search request and the E-commerce service auxiliary information.
13. An electronic device, comprising:
a memory storing an executable program;
a processor for executing the program, wherein the program executes the search processing method according to any one of claims 1 to 12 when executed.
14. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored executable program, wherein the executable program, when run, controls a device in which the computer-readable storage medium is located to perform the search processing method according to any one of claims 1 to 12.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311219863.3A CN117349515A (en) | 2023-09-20 | 2023-09-20 | Search processing method, electronic device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311219863.3A CN117349515A (en) | 2023-09-20 | 2023-09-20 | Search processing method, electronic device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117349515A true CN117349515A (en) | 2024-01-05 |
Family
ID=89367451
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311219863.3A Pending CN117349515A (en) | 2023-09-20 | 2023-09-20 | Search processing method, electronic device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117349515A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118095448A (en) * | 2024-04-26 | 2024-05-28 | 支付宝(杭州)信息技术有限公司 | Medical auxiliary information extraction method and related equipment for LLM model |
-
2023
- 2023-09-20 CN CN202311219863.3A patent/CN117349515A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118095448A (en) * | 2024-04-26 | 2024-05-28 | 支付宝(杭州)信息技术有限公司 | Medical auxiliary information extraction method and related equipment for LLM model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10255546B2 (en) | Question-answering system | |
CN117521675A (en) | Information processing method, device, equipment and storage medium based on large language model | |
Biondi et al. | A deep learning semantic approach to emotion recognition using the IBM watson bluemix alchemy language | |
CN113392265A (en) | Multimedia processing method, device and equipment | |
CN116662522B (en) | Question answer recommendation method, storage medium and electronic equipment | |
CN116541493A (en) | Interactive response method, device, equipment and storage medium based on intention recognition | |
US11907863B2 (en) | Natural language enrichment using action explanations | |
Hassani et al. | LVTIA: A new method for keyphrase extraction from scientific video lectures | |
CN117349515A (en) | Search processing method, electronic device and storage medium | |
CN118035945B (en) | Label recognition model processing method and related device | |
CN115270746A (en) | Question sample generation method and device, electronic equipment and storage medium | |
CN117932022A (en) | Intelligent question-answering method and device, electronic equipment and storage medium | |
CN117421413A (en) | Question-answer pair generation method and device and electronic equipment | |
CN117332035A (en) | Index generation method, electronic device, and storage medium | |
CN117473057A (en) | Question-answering processing method, system, equipment and storage medium | |
CN116701604A (en) | Question and answer corpus construction method and device, question and answer method, equipment and medium | |
CN113542797A (en) | Interaction method and device in video playing and computer readable storage medium | |
CN116186220A (en) | Information retrieval method, question and answer processing method, information retrieval device and system | |
CN113807920A (en) | Artificial intelligence based product recommendation method, device, equipment and storage medium | |
CN118227910B (en) | Media resource aggregation method, device, equipment and storage medium | |
CN118132729B (en) | Answer generation method and device based on medical knowledge graph | |
CN118296132B (en) | Customer service searching method and system based on intelligent large model | |
Xue et al. | An ontology-supported inquiry learning technique | |
Das et al. | Leveraging facial expressions as emotional context in image captioning | |
Sabharwal et al. | Various Cognitive Platforms/Engines |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |