CN112784141B - Search result quality determination method, apparatus, storage medium and computer device - Google Patents
Search result quality determination method, apparatus, storage medium and computer device Download PDFInfo
- Publication number
- CN112784141B CN112784141B CN201911011937.8A CN201911011937A CN112784141B CN 112784141 B CN112784141 B CN 112784141B CN 201911011937 A CN201911011937 A CN 201911011937A CN 112784141 B CN112784141 B CN 112784141B
- Authority
- CN
- China
- Prior art keywords
- quality
- search
- evaluation
- search result
- assessment
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000011156 evaluation Methods 0.000 claims abstract description 176
- 238000013441 quality evaluation Methods 0.000 claims abstract description 142
- 238000001303 quality assessment method Methods 0.000 claims description 42
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 8
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 238000013210 evaluation model Methods 0.000 description 39
- 238000010586 diagram Methods 0.000 description 14
- 238000002372 labelling Methods 0.000 description 11
- 238000004422 calculation algorithm Methods 0.000 description 5
- 206010028813 Nausea Diseases 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000008693 nausea Effects 0.000 description 4
- 241000220317 Rosa Species 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 244000241235 Citrullus lanatus Species 0.000 description 1
- 235000012828 Citrullus lanatus var citroides Nutrition 0.000 description 1
- 240000000716 Durio zibethinus Species 0.000 description 1
- 235000006025 Durio zibethinus Nutrition 0.000 description 1
- 108091081062 Repeated sequence (DNA) Proteins 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 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
- G06F16/9532—Query formulation
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to a search result quality determination method, a device, a storage medium and equipment, wherein the method comprises the steps of firstly obtaining a search result to be evaluated, determining a plurality of evaluation dimensions for performing quality evaluation on the search result, then inputting the search result into a plurality of quality evaluation models corresponding to the plurality of evaluation dimensions, enabling the plurality of quality evaluation models to perform quality evaluation processing under the corresponding evaluation dimensions on the search evaluation result respectively, obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results, and finally determining the quality of the search result based on the plurality of quality evaluation results. According to the scheme, the quality evaluation of the search results can be performed under different evaluation dimensions by utilizing a plurality of quality evaluation models, and the quality of the search results is finally determined by integrating the multi-dimensional evaluation results, so that the quality of the search results is improved, and the method is also beneficial to quickly and accurately hitting high-quality search results in massive search resources for the search models.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a method and apparatus for determining quality of a search result, a computer readable storage medium, and a computer device.
Background
With the rapid development of information processing technology, the internet provides a large amount of information resources for users to access, and users can access the internet through intelligent terminals such as mobile phones and tablet computers to obtain required information resources, such as inquiring and downloading the required information resources through search engines. The information resources are usually presented on the terminal in the form of search results, and the improvement of the quality of the search results is beneficial to the user to acquire the corresponding information resources more quickly, so that accurate evaluation basis is required to be provided for the quality of the search results.
The conventional technology mainly depends on the click rate of the user on the search result so as to evaluate the search result, and thus only the read condition of the search result can be reflected in a statistical sense, so that the evaluation on the quality of the search result is inaccurate.
Disclosure of Invention
Based on this, there is a need to provide a search result quality determination method, apparatus, computer-readable storage medium, and computer device for solving the technical problem that the conventional technology is inaccurate in search result quality evaluation.
A search result quality determination method, comprising:
acquiring a search result to be evaluated;
determining a plurality of evaluation dimensions for quality evaluation of the search results;
inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results;
obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results;
a quality of the search results is determined based on the plurality of quality assessment results.
A search result quality determination apparatus, the apparatus comprising:
the search result acquisition module is used for acquiring search results to be evaluated;
an evaluation dimension determination module for determining a plurality of evaluation dimensions for quality evaluation of the search results;
a search result input module for inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results;
the evaluation result acquisition module is used for acquiring the quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results;
And the search quality determining module is used for determining the quality of the search results based on the plurality of quality evaluation results.
A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
acquiring a search result to be evaluated; determining a plurality of evaluation dimensions for quality evaluation of the search results; inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results; obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results; a quality of the search results is determined based on the plurality of quality assessment results.
A computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
acquiring a search result to be evaluated; determining a plurality of evaluation dimensions for quality evaluation of the search results; inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results; obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results; a quality of the search results is determined based on the plurality of quality assessment results.
The method comprises the steps of firstly obtaining a search result to be evaluated, determining a plurality of evaluation dimensions for evaluating the quality of the search result, inputting the search result into a plurality of quality evaluation models corresponding to the evaluation dimensions, enabling the quality evaluation models to respectively perform quality evaluation processing on the search evaluation result under the corresponding evaluation dimensions, obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results, and finally determining the quality of the search result based on the quality evaluation results. According to the scheme, the quality evaluation of the search results can be performed under different evaluation dimensions by utilizing a plurality of quality evaluation models, and the quality of the search results is finally determined by integrating the multi-dimensional evaluation results, so that the quality of the search results is improved, and the method is also beneficial to quickly and accurately hitting high-quality search results in massive search resources for the search models.
Drawings
FIG. 1 is a diagram of an application environment for a search result quality determination method in one embodiment;
FIG. 2 is a schematic diagram of a terminal interface in a search scenario in one embodiment;
FIG. 3 is a flow diagram of a method for determining quality of search results in one embodiment;
FIG. 4 is a schematic diagram of a search result quality determination method in a search model in an application example;
FIG. 5 is a block diagram of a search result quality determination apparatus in one embodiment;
FIG. 6 is a block diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The method for determining the quality of the search result provided by the application can be applied to an application environment as shown in fig. 1, and fig. 1 is an application environment diagram of the method for determining the quality of the search result in one embodiment, where the application environment can include a terminal 101 and a server 102, and the terminal 101 can be connected with the server 102 through a network such as the internet. The server 102 may provide information resources for the terminal 101, the user may input a search term on the terminal 101, the terminal 101 sends the search term to the server 102 through a network, and the server 102 may feed back the corresponding information resources to the terminal 101 for display in the form of search results, so that the user may view the search results on the terminal 101 and obtain the corresponding information resources.
Referring to fig. 2 for a brief description of a search process, fig. 2 is a schematic diagram of a terminal interface in a search scenario in an embodiment, as shown in fig. 2, a user may open a search box 201 directly through a search on a terminal 101, input a search word in the search box 201, for example, the user may input "the small clear world rank" as the search word into the search box 201, then the server 102 may send a search result list 202 corresponding to the search word to the terminal 101 for displaying, the user may browse each search result in the search result list 202, for example, when the user needs to browse the search result 1, the server 102 may further send a content page corresponding to the search result 1 to the terminal 101 for displaying, where the content page may show contents including a title, a text, and the like of the search result 1, so that the user may obtain corresponding information resources on the terminal 101.
Therefore, the improvement of the quality of the search result provided by the server 102 is beneficial to the user to obtain the corresponding information resource on the terminal 101 more quickly, while the method for determining the quality of the search result provided by the server 102 can be applied to accurately evaluating the quality of the search result provided by the server 102, specifically, the server 102 can obtain the search result to be evaluated, determine a plurality of evaluation dimensions for evaluating the quality of the search result, input the search result into a plurality of quality evaluation models corresponding to the evaluation dimensions, so that the quality evaluation models respectively evaluate the quality of the search result in the corresponding dimensions, then the server 102 can obtain the quality evaluation results of the quality evaluation models, thereby obtaining a plurality of quality evaluation results, and finally the server 102 synthesizes the quality evaluation results to finally determine the quality of the search result.
In the above scenario, the terminal 101 may be a desktop terminal or a mobile terminal, and the mobile terminal may specifically be at least one of a mobile phone, a tablet computer, a notebook computer, and the like, and the server 102 may be implemented by a separate server or a server cluster formed by a plurality of servers.
In one embodiment, a method for determining quality of a search result is provided, and this embodiment is mainly described by applying the method to the server 102 in fig. 1, as shown in fig. 3, fig. 3 is a schematic flow diagram of the method for determining quality of a search result in one embodiment, where the method for determining quality of a search result may include the following steps:
s301, obtaining a search result to be evaluated.
In this step, the server 102 may extract, in a random sampling manner, search results stored in a local database of the server 102 as search results to be evaluated, where the search results generally correspond to a certain search term, and the number of search results corresponding to each search term is generally plural. For example, when a user searches for related information resources using "small Ming world rank" as a target search term on the terminal 101, the server 102 may obtain a plurality of search results corresponding to the target search term through a search model and feed the search results back to the terminal 101 for presentation, while the server 102 may store the search results and the corresponding target search term in a local database as historical search data, and when the quality of the search results provided by the server 102 needs to be evaluated, the server 102 may obtain one or more search results from the local database as the search results to be evaluated.
S302, determining a plurality of evaluation dimensions for evaluating the quality of the search results.
In this step, the server 102 may acquire multiple evaluation dimensions for evaluating the quality of the search result, where evaluating the quality of the search result refers to evaluating whether the search result is a good quality search result or a poor quality search result, and in this step, mainly from multiple dimensions, multiple dimensions for evaluating the quality of the search result are determined, so as to evaluate whether the search result is a good quality search result or a poor quality search result in different evaluation dimensions. The evaluation dimensions may be formulated according to actual needs, for example, the quality evaluation may be performed on the search results from evaluation dimensions such as a search result availability evaluation dimension, a search result relevance evaluation dimension, an intention satisfaction evaluation dimension, a content quality evaluation dimension, and a search result repeatability evaluation dimension, which may be combined or increased or decreased according to actual needs.
S303, inputting the search results to a plurality of quality evaluation models corresponding to a plurality of evaluation dimensions, respectively.
In this step, after determining a plurality of evaluation dimensions for performing quality evaluation on the search result, the server 102 may further obtain a plurality of quality evaluation models corresponding to the plurality of evaluation dimensions, where each quality evaluation model corresponds to a different evaluation dimension, so that the server 102 may input the search result to be evaluated into the plurality of quality evaluation models, respectively, so that the plurality of quality evaluation models perform quality evaluation under the corresponding evaluation dimensions on the search result to be evaluated, respectively. Wherein, these quality evaluation models may be pre-constructed by the server 102, the server 102 may store each pre-constructed quality evaluation model in an evaluation model library, and since each quality evaluation model corresponds to a different evaluation dimension, when the server 102 needs to evaluate the search result, after determining multiple evaluation dimensions, a corresponding multiple quality evaluation models may be extracted from the evaluation model library, and then the search result may be input to these quality evaluation models, so that these quality evaluation models may correspond to the quality evaluation under the evaluation dimension to the search result.
S304, quality evaluation results of the quality evaluation models are obtained, and a plurality of quality evaluation results are obtained.
In this step, after the server 102 inputs the search result into the quality assessment models, the quality assessment models perform quality assessment processing on the search result under the corresponding assessment dimensions, so as to output quality assessment results of the corresponding assessment dimensions, and the server 102 may obtain the quality assessment results output by the quality assessment models, so as to obtain quality assessment results corresponding to different assessment dimensions.
S305, determining the quality of the search result based on a plurality of quality evaluation results.
In this step, the server 102 may integrate the quality evaluation results of the multiple evaluation dimensions to finally determine the quality of the search result. The quality evaluation result of each evaluation dimension may be embodied in the form of a quality evaluation score, that is, the server 102 may calculate an overall quality score of the search result based on the quality evaluation score of each evaluation dimension, so as to determine the quality of the search result according to the overall quality score. For example, the quality evaluation scores of all the evaluation dimensions can be combined with preset weight values to carry out weighted summation, so that the overall quality score of the search result is obtained, and the quality of the search result is evaluated according to the overall quality score. By adopting the method, the quality evaluation of the search results corresponding to the target search words can be obtained, various low-quality results in a complex search scene can be systematically and flowthrough found, a system framework is provided for the search model of the server 102 to quickly and accurately hit high-quality corresponding results in massive search resources, and data reference can be provided for an optimal approximate ordering algorithm.
According to the method for determining the quality of the search results, the server firstly acquires the search results to be evaluated, determines a plurality of evaluation dimensions for performing quality evaluation on the search results, then inputs the search results into a plurality of quality evaluation models corresponding to the evaluation dimensions, so that the quality evaluation models respectively perform quality evaluation processing on the search evaluation results under the corresponding evaluation dimensions, then acquires the quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results, and finally determines the quality of the search results based on the quality evaluation results. According to the scheme, the quality evaluation of the search results can be performed under different evaluation dimensions by utilizing a plurality of quality evaluation models, and the quality of the search results is finally determined by integrating the multi-dimensional evaluation results, so that the quality of the search results is improved, and the method is also beneficial to quickly and accurately hitting high-quality search results in massive search resources for the search models.
In one embodiment, the method for determining quality of search results may further include the following steps of constructing a quality evaluation model, and specifically includes:
acquiring resource quality characteristics of search service resources; determining the evaluation content of each evaluation dimension; generating a quality assessment scheme of each assessment dimension according to the assessment content and the resource quality characteristics; and constructing a quality evaluation model corresponding to each evaluation dimension based on the quality evaluation scheme.
In this embodiment, the server 102 may obtain the resource quality feature of the search service resource, where the search service resource refers to various information resources that can be searched when the server 102 provides the search service, for example, information resources such as videos, novels, pictures, etc., where the resource quality feature is mainly used to characterize the quality of the search service resource and features possessed by the inferior search service resource, for example, the superior video resource may generally provide a stable video access page and a title that accurately describes the video content, the superior novels resource may generally provide a novel access page that is normative of text typesetting, a title that accurately describes the novel content, etc., and for the features of the inferior search service resource, the page content corresponding to the search result is usually not identical to the search result title, the search result includes violent and three-custom content, etc., based on this, the server 102 may collect the resource quality feature of the search service resource of various qualities, and also need to determine the evaluation content of various evaluation dimensions, for example, for the search result availability evaluation dimensions, whether the title of the search result may be used as evaluation content, and if the search result title may be used as evaluation content, the evaluation dimension may be generated according to the evaluation model of the quality of the search result, and the evaluation may be based on the evaluation of the quality of the corresponding to the evaluation of the search result, and the evaluation of the evaluation results.
The above embodiment provides a quality evaluation model construction scheme corresponding to each evaluation dimension, and can determine evaluation contents of multiple evaluation dimensions from the dimensions such as the title quality, the content quality and the overall user experience of the search service resource by summarizing the features of the high-quality and low-quality resources of the search service, and construct a quality evaluation model corresponding to the dimensions including the availability of the search results, the relevance of the search results, the satisfaction of the intention, the quality of the content and the repetition of the search results.
The respective quality assessment models are described below from a search result availability assessment dimension, a search result relevance assessment dimension, an intent satisfaction assessment dimension, a content quality assessment dimension, and a search result repeatability assessment dimension, respectively.
In one embodiment, the quality evaluation model corresponding to the search result availability evaluation dimension may include an availability evaluation model, where the availability evaluation model is mainly used to determine a sensitive content type of the search result, and obtain the availability evaluation result of the search result according to the sensitive content type.
In this embodiment, the server 102 may obtain an availability evaluation result of the search result through an availability evaluation model, where the availability evaluation result is a quality evaluation result of the search result in a search result availability evaluation dimension.
In this step, the server 102 may input the search result into the availability evaluation model, and after the availability evaluation model obtains the search result, the sensitive content type of the search result may be determined, and the availability evaluation result of the search result may be obtained based on the sensitive content type. The availability evaluation model may determine whether the search result contains sensitive content, and thus determine the type of the sensitive content, and the server 102 may allocate different availability evaluation scores to different types of sensitive content in advance, so that after the server 102 determines the type of the sensitive content of the search result, the availability evaluation score corresponding to the type of the sensitive content may be further queried, and the availability evaluation score may be used as the availability evaluation result of the search result. For example, the usability evaluation model may determine whether the search result contains sensitive content such as pornography, hypo-colloquiality, nausea, politics sensitive events, and the like, then determine the type of sensitive content such as pornography corresponding to sensitive content type a, hypo-colloquiality corresponding to sensitive content type B, nausea corresponding to sensitive content type C, and the like according to whether the search result contains the sensitive content, and then query the corresponding usability evaluation score according to the sensitive content type, such as the sensitive content type a corresponding evaluation score is a, the sensitive content type B corresponding evaluation score is B, and the like, so that the corresponding usability evaluation score may be used as the usability evaluation result.
In some embodiments, the usability evaluation model may be further configured to obtain a search result title of the search result, obtain a preset sensitive word, and determine a sensitive content type of the search result according to the sensitive word and the search result title.
In this embodiment, the usability evaluation model determines the type of the sensitive content of the search result according to whether the search result title of the search result contains a preset sensitive word. The preset sensitive words may include a plurality of sensitive words, different sensitive words may correspond to different sensitive content types, the usability evaluation model may identify the sensitive words contained in the search result titles to determine the sensitive content types of the search results, and may further label the sensitive content types to which the search results belong, so as to facilitate the subsequent classification processing of the search results, where the usability evaluation model may specifically refer to the following table 1 and table 2 to perform quality evaluation and labeling processing on the search result titles of the search results, table 1 is a scoring example table of whether the search result titles contain "pornography, suborder, nausea content", and table 2 is a scoring example table of whether the search result titles contain political sensitive events.
TABLE 1
TABLE 2
In one embodiment, the quality assessment model corresponding to the search result relevance assessment dimension may include a relevance assessment model that may be used to determine a relevance assessment result for a search result based on a degree of matching of a search result title of the search result to a target search term. Where the search result title refers to the title of the search result and the target search term is the search term corresponding to the search result.
In this embodiment, the relevance evaluation model may obtain a first core word of a target search word and a second core word of a search result title, where the core word refers to a word or phrase that can most express a meaning of the core word, and generally includes a proper noun, a physical verb, a necessary binode-like complement component thereof, and the like, and then the relevance evaluation model may be used to determine a matching type of the target search word and the search result title according to the first core word and the second core word, where the relevance evaluation model may firstly cut the search word and the title into a plurality of words having actual meanings through word cutting, calculate a weight of each word, thereby obtaining a corresponding core word, and then calculate a matching type of the title and the search word corresponding to the search result title by combining a coverage rate of the core word and the title, a semantic vector similarity, and the like, of the core word and the search word; finally, the relevance evaluation model may obtain a relevance evaluation result of the search result according to the matching type. The correlation evaluation model may set different matching types for the different matching degrees between the target search word and the search result title in advance, and may further set different evaluation values for the different matching types, so that the correlation evaluation model may query the corresponding evaluation values as correlation evaluation results after determining the matching types, and in some embodiments, the correlation evaluation model may specifically refer to the following table 3 to obtain the correlation evaluation results of the search results, and table 3 is a correlation evaluation score comparison table of the search results.
TABLE 3 Table 3
In one embodiment, the quality evaluation model corresponding to the intent satisfaction evaluation dimension may include a satisfaction evaluation model, where the satisfaction evaluation model is mainly used for obtaining a target search intent corresponding to a target search word, obtaining a search result title of a search result, determining an intent satisfaction type of the search result according to the search result title and the target search intent, and obtaining an intent satisfaction evaluation result of the search result according to the intent satisfaction type.
According to the method, whether the search result can intuitively meet the original search intention of a user is measured based on the satisfaction evaluation model, the original search intention can be judged by acquiring a head result of a main stream search engine, which is not an advertisement, and an intention analysis model is formed through data accumulation, so that the intention of similar search words is calculated based on the intention analysis model, the intention of similar search words is deduced, for example, the head result of the search engine for "eating watermelon belly pain" is an analysis cause, a suggested medical result is given, and an intention analysis model is manufactured by accumulating related data, so that the intention of "eating durian belly pain" is deduced to be similar intention: the etiology and the solution. Therefore, after obtaining the target search word, the satisfaction evaluation model may further obtain the target search intent corresponding to the target search word, where the target search word intent may be pre-stored in a database local to the server 102, and when the intent satisfaction needs to be evaluated, the corresponding search intent is queried according to the target search word as the target search intent, and the satisfaction evaluation model may further need to obtain a search result title of the search result, so that the intent satisfaction type corresponding to the search result may be further determined according to the intent matching degree of the search result title and the target search intent, the corresponding intent satisfaction evaluation score may be set for different intent matching degrees and different intent satisfaction types, so that the satisfaction evaluation score may be queried for different intent satisfaction types after the satisfaction evaluation model determines the intent satisfaction type, and in some embodiments, the satisfaction evaluation model may specifically obtain the intent satisfaction evaluation score corresponding to the non-agreeable pattern satisfaction type with reference to the following table 4, and the table 4 is the intent satisfaction evaluation score corresponding to the intent satisfaction table.
TABLE 4 Table 4
In one embodiment, the number of search results corresponding to the target search term may be plural, and the poor ranking in each search result may be further found by the following steps:
and obtaining the sequencing results of the search results, obtaining the intention satisfaction degree evaluation results of the search results, and determining poor sequencing in the sequencing results based on the intention satisfaction degree evaluation results of the search results.
In this embodiment, the server 102 first obtains the ranking results of the plurality of search results corresponding to the target search term, and then may obtain the intended satisfaction evaluation results of the search results through the satisfaction evaluation model of the above embodiment, where table 5 is a search result rank and intended satisfaction comparison table, and the intended satisfaction evaluation results of the search results may be represented by the intended satisfaction evaluation score.
| Result rank order | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Degree of satisfaction of intention | 3 | 2 | 3 | 2 | 2 | 1 | 2 | 1 | 0 | -1 |
TABLE 5
The ranking results of the first 10 search results corresponding to the target search term are displayed in the result rank, and the intent satisfaction is the intent satisfaction evaluation score corresponding to the 10 search results, and based on the table 5, it can be found that the following situations are included, where the intent satisfaction evaluation score is the same: search results with order of results 1 and 3 were rated for 3 points, search results with order of 2, 4, 5 and 7 were rated for 2 points, and search results with order of 6 and 8 were rated for 1 point. The ranking results are set to meet the principle of decreasing intention satisfaction degree, so that the score is evaluated for 3-point intention satisfaction degree, but the search result with the result rank of 1 st in terms of meeting the original search requirement should be obviously superior to the 3 rd search result in subjective feeling, i.e. the user can click on the 1 st search result preferentially. Then, in the case where the degree rank sequence is intended to be satisfied, lower case letters may be marked (e.g., may be marked starting from lower case letter a) at the 1 st and 3 rd order results, and for the case where the degree is intended to be satisfied 2 nd, there is also a case where the 7 th order result is better than the 4 th order result, then it is necessary to mark lower case letters (e.g., lower case letter b) at each of the 4 th, 5 th and 7 th, at position 2, no marks are needed, as the ordering is reasonable, it will be appreciated that there may be multiple sets of ordering results intended to meet the degree, and therefore different lowercase letters need to be used to mark, so that poor ordering in the ordering results can be determined from the marks.
In one embodiment, the quality assessment model corresponding to the quality assessment dimension may include a quality assessment model, where the quality assessment model is mainly used for performing quality assessment on page content corresponding to the search result. The quality evaluation model can acquire page contents corresponding to the search results, determine a plurality of page content evaluation items, acquire a plurality of page content evaluation results of the page contents under the page content evaluation items, and acquire the quality evaluation results of the search results according to the page content evaluation results.
In this embodiment, the server 102 may utilize the merit evaluation model to perform quality evaluation under the merit evaluation dimension on the page content corresponding to the search result, and measure whether the search result after being clicked by the user can meet the user expectation. Wherein the server 102 may preset a plurality of page content evaluation items, which may include, but are not limited to, in some embodiments: content accessibility assessment, content correctness assessment, content richness assessment, content authority assessment, content integrity assessment, content timeliness assessment, and content usability assessment. The server 102 may perform a quality assessment on the page content corresponding to the search result based on the page content assessment items through the quality assessment model, obtain a plurality of page content assessment results of the page content under a plurality of page content assessment items, where the page content assessment results may be content assessment scores under different page content assessment items, and finally calculate the quality assessment result of the search result according to the content assessment scores. In a specific application, the merit evaluation model may refer to the following tables 6 to 12 to evaluate the page content corresponding to the search result under each page content evaluation item, and obtain the corresponding score to obtain the merit evaluation result of the search result, where tables 6 to 12 are a content accessibility evaluation comparison table, a content correctness evaluation comparison table, a content richness evaluation comparison table, a content authority evaluation comparison table, a content integrity evaluation comparison table, a content timeliness evaluation comparison table, and a content usability evaluation comparison table, respectively.
TABLE 6
TABLE 7
/>
TABLE 8
TABLE 9
/>
Table 10
TABLE 11
Table 12
In one embodiment, the quality evaluation model corresponding to the search result repetition evaluation dimension may include a repetition evaluation model, where the repetition evaluation model is mainly used to obtain other search results corresponding to the target search word, determine a search result repetition type between the search result and the other search results, and obtain a repetition evaluation result of the search result according to the search result repetition type.
In this embodiment, the server 102 may determine, through a repeatability evaluation model, a repetition type between a certain search result corresponding to a target search word and other search results, and obtain, according to the repetition type to which the search result belongs, a repeatability evaluation result of the search result, where the repeatability evaluation result may be identified by using a case-letter english letter.
For example, in all search results for the same search term, there may be duplicate types:
(a) The search result titles of the plurality of search results are normalized and have high similarity, and the search result titles are in the control of being shorter, and have more than three fourths of characters, for example, shenzhen subway-encyclopedia Shenzhen Metro-encyclopedia Shenzhen subway-encyclopedia, the situation can be evaluated without considering the "encyclopedia" behind the title, because the "encyclopedia" can influence the numerical value of similarity calculation;
(b) The plurality of search result contents are basically consistent, and the main contents meet the same requirement or describe the same things, such as what the rose meaning and what the rose meaning;
(c) Multiple search results meet the same requirement, e.g., QQ browser & QQ browser very fast version & QQ browser VR version;
(d) Two or more search results are identical, such as APEX-translation & APEX-translation;
(e) The landing pages are completely consistent after two or more search results are clicked;
(f) And (5) after the cover pictures of the two search results are scaled, judging that the results are repeated if the cover pictures are completely consistent (such as information short videos).
In the above embodiment, there may be multiple repetition types in the same batch of search results corresponding to the target search word, each repetition type may correspond to one repetition group, each repetition group may be marked with the same lowercase letter and be distinguished from other repetition groups, and for the openness problem, and the answer content in the search results is different and may not be considered as repetition.
In one embodiment, the number of search results may be plural, and after determining the quality of the search results based on the plural quality evaluation results in step S305, the steps of:
And acquiring the quality of each search result, and determining the search quality of the target search word according to the quality of each search result.
In this embodiment, the server 102 may obtain the quality of multiple search results corresponding to the same target search term, and synthesize the quality of the search results to obtain the search quality of the target search term, for example, may perform processing methods such as weighted average or weighted summation on the quality of the search results of different orders, obtain the search quality score of the target search term, determine the search quality of the target search term according to the search quality score, where the search quality of the target search term may provide a data reference for a related search engine, for example, provide a data reference for an optimal approximation ranking algorithm, and further improve the search model thereof.
In order to more clearly illustrate the technical scheme of the search result quality determination method provided by the embodiment of the application, the method is applied to search result quality evaluation on a multi-layer search model in combination with fig. 4, wherein fig. 4 is a schematic diagram of the search result quality determination method in the search model in an application example, a user can send a search word to a server 102 through a terminal 101, the server 102 processes the search word layer by layer through the multi-layer search model, wherein a resource layer corresponds to an abstract level of a database in which the search result is stored, a recall layer extracts all relevant results through a recall algorithm, a coarse ranking layer and a fine ranking layer are subjected to two-time large-scale ranking through a comprehensive ranking algorithm, and finally a display control layer is responsible for processing such as de-duplication and style display, and the search result is also fed back to the terminal 101 for display. The search result quality determining method provided by the embodiment of the application can be applied to quality evaluation processing including content quality, relevance, availability, timeliness, intention satisfaction, repeatability and the like on the layers, systematically finds out bad results in the search models of the layers of the upper graph, reduces subjective errors of manual judgment through complete and independent evaluation interval division, clearly locates the result most conforming to the user expectation through complete and independent evaluation dimensions and corresponding intervals, is not limited by a statistical representation method with high clicking and high residence, and can promote the search result to develop to a more humanized and more conforming to the benign environment expected by the user.
The specific evaluation flow is as follows:
1. usability assessment: the method can judge whether the titles and the thumbnail information of all the search results exist in the current search results: (1) Pornography, hypo, nausea, single scoring, labeling [ 0,1 ] [ a, B, C, D ], respectively; (2) Political sensitive events, single scoring, labeling [ 0,1 ] respectively;
2. correlation evaluation: the text relativity (including obvious semantic relativity) of the titles of all search results and search words in the search results can be judged, and corresponding results [ 1,0,1,2,3 ] are marked;
3. intent satisfaction evaluation: the intention satisfaction degree of all search results on the search words (the search results of the reference association words and the search engine) in the search results can be judged, and corresponding results [ 1,0,1,2,3 ] are marked;
4. content quality assessment: evaluating the quality of the resource content corresponding to the search result comprises the following steps:
(1) Content accessible, single scoring, labeling [ 0,1 ] [ A, B, C, D, E ]
(2) Content correctness, single scoring, labeling [ 0,1 ] [ A, B, C, D, E ]
(3) Content richness, single scoring, labeling [ 1,0,1 ]
(4) Content authority, single scoring, labeling [ 1,0,1 ]
(5) Content integrity, single scoring, labeling [ 1,0,1 ] [ A, B, C, D, E, F ]
(6) Content timeliness, single scoring, labeling [ 0,1 ]
(7) Internally easy availability, single scoring, labeling [ 0,1 ]
(8) Content quality, comprehensive scoring, labeling [ 1,0,1,2,3 ]
5. Search results repeat evaluation: comprehensively looking at all search results in the search results, marking all repeated search results according to the repeated sequences: e.g., results repeat 1 and 3, results repeat 2, 5, and 7, then label [ a ] at 1 and 3, and label [ b ] at 2, 5, and 7.
The quality evaluation results of the finally obtained search results are exemplified as follows:
according to the evaluation flow for evaluating the quality of the search results in the application example, the quality evaluation model based on the usability, the relativity, the intention satisfaction, the content quality and the search results, which is repeated in five dimensions, can be designed by summarizing the characteristics of the search service, namely the quality of the inferior resources, and starting from the three dimensions of the title quality, the content quality and the overall experience of the resources, so that the overall quality score of the search contents can be obtained, the search results can be ranked according to the quality score, systematic and procedural discovery of various low-quality results in complex search scenes can be facilitated, a system framework is provided for the search model to rapidly and accurately hit the corresponding results with high quality in mass search resources, and data reference is provided for the optimal approximate ranking algorithm, and meanwhile, the result which is most in line with the expectations of the users can be clearly positioned through the complete and independent evaluation dimensions and the corresponding regions, and the statistical representation method which is not limited by high click and high stay can promote the search results to develop more humanized and more in line with the benign environment expected by the users.
In one embodiment, a search result quality determining apparatus is provided, as shown in fig. 5, fig. 5 is a block diagram of a search result quality determining apparatus in one embodiment, and the search result quality determining apparatus 500 may include:
a search result obtaining module 501, configured to obtain a search result to be evaluated;
an evaluation dimension determination module 502 for determining a plurality of evaluation dimensions for quality evaluation of search results;
a search result input module 503 for inputting search results to a plurality of quality evaluation models corresponding to a plurality of evaluation dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results;
an evaluation result obtaining module 504, configured to obtain quality evaluation results of the plurality of quality evaluation models, and obtain a plurality of quality evaluation results;
the search quality determination module 505 is configured to determine a quality of a search result based on a plurality of quality evaluation results.
In one embodiment, the method may further include: the model building module is used for acquiring the resource quality characteristics of the search service resources; determining the evaluation content of each evaluation dimension; generating a quality assessment scheme of each assessment dimension according to the assessment content and the resource quality characteristics; and constructing a quality evaluation model corresponding to each evaluation dimension based on the quality evaluation scheme.
In one embodiment, evaluating the dimensions may include: a search result availability evaluation dimension, a search result relevance evaluation dimension, an intent satisfaction evaluation dimension, a content quality evaluation dimension, and/or a search result repeatability evaluation dimension.
In one embodiment, the quality evaluation model corresponding to the search result availability evaluation dimension may include an availability evaluation model for determining a sensitive content type of the search result, and obtaining the availability evaluation result of the search result according to the sensitive content type.
In one embodiment, the usability evaluation model is further configured to obtain a search result title of the search result, obtain a preset sensitive word, and determine a sensitive content type of the search result according to the sensitive word and the search result title.
In one embodiment, the quality evaluation model corresponding to the search result relevance evaluation dimension may include a relevance evaluation model, configured to obtain a first core word of a target search word and a second core word of a search result title, determine a matching type of the target search word and the search result title according to the first core word and the second core word, and obtain a relevance evaluation result of the search result according to the matching type; the search result title is the title of the search result; the target search term is a search term corresponding to the search result.
In one embodiment, the quality evaluation model corresponding to the intention satisfaction evaluation dimension may include a satisfaction evaluation model for acquiring a target search intention corresponding to a target search word, acquiring a search result title of a search result, determining an intention satisfaction type of the search result according to the search result title and the target search intention, and acquiring an intention satisfaction evaluation result of the search result according to the intention satisfaction type.
In one embodiment, the number of search results is a plurality; the search system also comprises a poor ranking result determining module, a poor ranking result determining module and a ranking module, wherein the poor ranking result determining module is used for acquiring ranking results of all search results; acquiring the intention satisfaction evaluation result of each search result; and determining poor ranking in the ranking results based on the intention satisfaction evaluation results of the search results.
In one embodiment, the quality evaluation model corresponding to the quality evaluation dimension may include a quality evaluation model for acquiring page content corresponding to the search result, determining a plurality of page content evaluation items, acquiring a plurality of page content evaluation results of the page content under the plurality of page content evaluation items, and acquiring the quality evaluation result of the search result according to the plurality of page content evaluation results.
In one embodiment, the page content evaluation item may include: content accessibility assessment, content correctness assessment, content richness assessment, content authority assessment, content integrity assessment, content timeliness assessment, and/or content usability assessment.
In one embodiment, the quality evaluation model corresponding to the search result repetition evaluation dimension may include a repetition evaluation model for acquiring other search results corresponding to the target search word, determining a search result repetition type between the search result and the other search results, and acquiring a repetition evaluation result of the search result according to the search result repetition type.
In one embodiment, the number of search results is a plurality; the method can also comprise a search word search quality determining module, which is used for obtaining the quality of each search result and determining the search quality of the target search word according to the quality of each search result.
FIG. 6 illustrates an internal block diagram of a computer device in one embodiment. The computer device may be specifically the server 102 of fig. 1. As shown in fig. 6, fig. 6 is a block diagram of a computer device including a processor, memory, and network interface connected by a system bus, in one embodiment. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program that, when executed by a processor, causes the processor to implement a search result quality determination method. The internal memory may also have stored therein a computer program which, when executed by the processor, causes the processor to perform a method of determining the quality of search results.
It will be appreciated by those skilled in the art that the structure shown in FIG. 6 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided that includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the search result quality determination method described above. The steps of the search result quality determination method herein may be the steps of the search result quality determination method of the above-described respective embodiments.
In one embodiment, a computer readable storage medium is provided, storing a computer program which, when executed by a processor, causes the processor to perform the steps of the search result quality determination method described above. The steps of the search result quality determination method herein may be the steps of the search result quality determination method of the above-described respective embodiments.
Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a non-volatile computer readable storage medium, and where the program, when executed, may include processes in the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (14)
1. A search result quality determination method, comprising:
acquiring resource quality characteristics of search service resources;
determining the evaluation content of each evaluation dimension;
generating a quality assessment scheme of each assessment dimension according to the assessment content and the resource quality characteristics;
constructing the quality evaluation model corresponding to each evaluation dimension based on the quality evaluation scheme;
Acquiring a search result to be evaluated;
determining a plurality of evaluation dimensions for quality evaluation of the search results;
inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results;
obtaining quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results;
a quality of the search results is determined based on the plurality of quality assessment results.
2. The method of claim 1, wherein the evaluating the dimension comprises: a search result availability evaluation dimension, a search result relevance evaluation dimension, an intent satisfaction evaluation dimension, a content quality evaluation dimension, and/or a search result repeatability evaluation dimension.
3. The method of claim 2, wherein the quality assessment model corresponding to the search result availability assessment dimension includes an availability assessment model for determining a sensitive content type of the search result, and obtaining an availability assessment result of the search result according to the sensitive content type.
4. The method of claim 3, wherein the usability assessment model is further configured to obtain a search result title of the search result, obtain a preset sensitive word, and determine a sensitive content type of the search result according to the sensitive word and the search result title.
5. The method according to claim 2, wherein the quality assessment model corresponding to the search result relevance assessment dimension includes a relevance assessment model for acquiring a first core word of a target search word and a second core word of the search result title, determining a matching type of the target search word and the search result title according to the first core word and the second core word, and acquiring a relevance assessment result of the search result according to the matching type; wherein the search result title is the title of the search result; the target search word is a search word corresponding to the search result.
6. The method of claim 2, wherein the quality assessment model corresponding to the intent satisfaction assessment dimension comprises a satisfaction assessment model for obtaining a target search intent corresponding to a target search term, obtaining a search result title of the search result, determining an intent satisfaction type of the search result based on the search result title and target search intent, and obtaining an intent satisfaction assessment result of the search result based on the intent satisfaction type.
7. The method of claim 6, wherein the number of search results is a plurality; further comprises:
obtaining the sequencing result of each search result;
acquiring the intention satisfaction degree evaluation result of each search result;
and determining bad orders in the ordering results based on the intention satisfaction degree evaluation results of the search results.
8. The method of claim 2, wherein the quality assessment model corresponding to the quality assessment dimension includes a quality assessment model for obtaining page content corresponding to the search result, determining a plurality of page content assessment items, obtaining a plurality of page content assessment results of the page content under the plurality of page content assessment items, and obtaining a quality assessment result of the search result according to the plurality of page content assessment results.
9. The method of claim 8, wherein the page content rating item comprises: content accessibility assessment, content correctness assessment, content richness assessment, content authority assessment, content integrity assessment, content timeliness assessment, and/or content usability assessment.
10. The method of claim 2, wherein the quality assessment model corresponding to the search result repetition assessment dimension comprises a repetition assessment model for obtaining other search results corresponding to a target search term, determining a search result repetition type between the search result and the other search results, and obtaining the repetition assessment result of the search result according to the search result repetition type.
11. The method of claim 1, wherein the number of search results is a plurality; after said determining the quality of the search results based on the plurality of quality assessment results, further comprising:
acquiring the quality of each search result;
and determining the search quality of the target search word according to the quality of each search result.
12. A search result quality determination apparatus, the apparatus comprising:
the model building module is used for acquiring the resource quality characteristics of the search service resources; determining the evaluation content of each evaluation dimension; generating a quality assessment scheme of each assessment dimension according to the assessment content and the resource quality characteristics; constructing the quality evaluation model corresponding to each evaluation dimension based on the quality evaluation scheme;
the search result acquisition module is used for acquiring search results to be evaluated;
an evaluation dimension determination module for determining a plurality of evaluation dimensions for quality evaluation of the search results;
a search result input module for inputting the search results to a plurality of quality assessment models corresponding to the plurality of assessment dimensions, respectively; the quality evaluation models are respectively used for performing quality evaluation under corresponding evaluation dimensions on the search results;
The evaluation result acquisition module is used for acquiring the quality evaluation results of the quality evaluation models to obtain a plurality of quality evaluation results;
and the search quality determining module is used for determining the quality of the search results based on the plurality of quality evaluation results.
13. A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method of any one of claims 1 to 11.
14. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method of any of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911011937.8A CN112784141B (en) | 2019-10-23 | 2019-10-23 | Search result quality determination method, apparatus, storage medium and computer device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911011937.8A CN112784141B (en) | 2019-10-23 | 2019-10-23 | Search result quality determination method, apparatus, storage medium and computer device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112784141A CN112784141A (en) | 2021-05-11 |
| CN112784141B true CN112784141B (en) | 2023-10-31 |
Family
ID=75747217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911011937.8A Active CN112784141B (en) | 2019-10-23 | 2019-10-23 | Search result quality determination method, apparatus, storage medium and computer device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112784141B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113254824B (en) * | 2021-05-14 | 2024-04-19 | 北京百度网讯科技有限公司 | Content determination method, device, medium, and program product |
| CN113449212B (en) * | 2021-06-25 | 2024-05-17 | 北京百度网讯科技有限公司 | Quality evaluation and optimization method, device and equipment for search results |
| CN113434775B (en) * | 2021-07-15 | 2024-03-26 | 北京达佳互联信息技术有限公司 | Method and device for determining search content |
| CN116680481B (en) * | 2023-08-03 | 2024-01-12 | 腾讯科技（深圳）有限公司 | Search ranking method, apparatus, device, storage medium and computer program product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103685189A (en) * | 2012-09-17 | 2014-03-26 | 百度在线网络技术（北京）有限公司 | Website security evaluation method and system |
| CN106570197A (en) * | 2016-11-15 | 2017-04-19 | 北京百度网讯科技有限公司 | Searching and ordering method and device based on transfer learning |
| CN106649760A (en) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | Question type search work searching method and question type search work searching device based on deep questions and answers |
| CN107704467A (en) * | 2016-08-09 | 2018-02-16 | 百度在线网络技术（北京）有限公司 | Search quality appraisal procedure and device |
-
2019
- 2019-10-23 CN CN201911011937.8A patent/CN112784141B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103685189A (en) * | 2012-09-17 | 2014-03-26 | 百度在线网络技术（北京）有限公司 | Website security evaluation method and system |
| CN107704467A (en) * | 2016-08-09 | 2018-02-16 | 百度在线网络技术（北京）有限公司 | Search quality appraisal procedure and device |
| CN106570197A (en) * | 2016-11-15 | 2017-04-19 | 北京百度网讯科技有限公司 | Searching and ordering method and device based on transfer learning |
| CN106649760A (en) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | Question type search work searching method and question type search work searching device based on deep questions and answers |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112784141A (en) | 2021-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112784141B (en) | Search result quality determination method, apparatus, storage medium and computer device | |
| CN109086439B (en) | Information recommendation method and device | |
| WO2020147720A1 (en) | Information recommendation method and device, and storage medium | |
| WO2017024884A1 (en) | Search intention identification method and device | |
| CN110888990B (en) | Text recommendation method, device, equipment and medium | |
| JP5575902B2 (en) | Information retrieval based on query semantic patterns | |
| CN102368262B (en) | Method and equipment for providing searching suggestions corresponding to query sequence | |
| US20150278359A1 (en) | Method and apparatus for generating a recommendation page | |
| CN110909182B (en) | Multimedia resource searching method, device, computer equipment and storage medium | |
| US10402479B2 (en) | Method, server, browser, and system for recommending text information | |
| US20080077569A1 (en) | Integrated Search Service System and Method | |
| US20160019219A1 (en) | Search result ranker | |
| CN109753601B (en) | Method and device for determining click rate of recommended information and electronic equipment | |
| CN106682049B (en) | Topic display system and topic display method | |
| CN110175264A (en) | Construction method, server and the computer readable storage medium of video user portrait | |
| CN111159563A (en) | Method, device and equipment for determining user interest point information and storage medium | |
| CN111753194B (en) | Information pushing method and device, electronic equipment and storage medium | |
| JP2017045196A (en) | Ambiguity evaluation device, ambiguity evaluation method, and ambiguity evaluation program | |
| CN111061954A (en) | Search result sorting method and device and storage medium | |
| CN112579854A (en) | Information processing method, device, equipment and storage medium | |
| JP2015106347A (en) | Recommendation device and recommendation method | |
| CN112597274A (en) | Document determination method, device, equipment and storage medium based on BM25 algorithm | |
| JP2014085862A (en) | Prediction server, program, and method for predicting number of future comments on prediction target content | |
| CN111651666A (en) | User theme recommendation method and device, computer equipment and storage medium | |
| CN109933709B (en) | Public opinion tracking method and device for video text combined data and computer equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |