Data processing method, device, apparatus, and computer-readable storage medium

Abstract

The present application discloses a data processing method, apparatus, device, and computer-readable storage medium, wherein the method comprises: obtaining an alternative document set corresponding to a user's query information; analyzing each alternative document in the alternative document set, determining a subject phrase corresponding to each alternative document, and obtaining a subject phrase set corresponding to the alternative document set; determining relevant information of the query information for each alternative document in the alternative document set, and obtaining a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determining a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, so as to improve the matching degree between the recalled document determined according to the query information and the query information.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, apparatus, device, and computer-readable storage medium. Background Art

[0002] At present, with the development of computer technology, search engines have become the main way for people to obtain information. When users use search engines to query information, they usually rely on the overlap between the user's query information and the words in the document content and the number of repeated words to directly determine the matching degree between the query information and the document, and then determine the recalled documents corresponding to the query information. The method of determining the matching degree between the query information and the document is relatively simple, and the matching degree between the determined recalled documents and the query information is low. Summary of the Invention

[0003] The embodiment of the present application provides an implementation solution different from the prior art to solve the technical problem that the recall document determined by the prior art has a low matching degree with the query information.

[0004] In a first aspect, the present application provides a data processing method, comprising:

[0005] Obtain the candidate document set corresponding to the user's query information;

[0006] Analyzing each candidate document in the candidate document set, determining a subject phrase group corresponding to each candidate document, and obtaining a subject phrase group group corresponding to the candidate document set;

[0007] Determining relevant information of the query information for each candidate document in the candidate document set, and obtaining a relevant information set of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, and a similarity value with the query information is greater than a preset similarity value;

[0008] A target recall result corresponding to the query information is determined based on the candidate document set, the subject phrase set, and the related information set.

[0009] In a second aspect, the present application provides a data processing device, comprising: an acquisition module, an analysis module, a first determination module, and a second determination module; wherein:

[0010] The acquisition module is used to obtain the candidate document set corresponding to the user's query information;

[0011] An analysis module, configured to analyze each candidate document in the candidate document set, determine a subject phrase group corresponding to each candidate document, and obtain a subject phrase group group corresponding to the candidate document set;

[0012] a first determining module configured to determine relevant information of the query information for each candidate document in the candidate document set, and obtain a set of relevant information of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, the query items having a similarity value with the query information greater than a preset similarity value;

[0013] The second determining module is configured to determine a target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set.

[0014] In a third aspect, the present application provides an electronic device, comprising:

[0015] processor; and

[0016] a memory for storing executable instructions for the processor;

[0017] The processor is configured to execute any method in the first aspect or any possible implementation of the first aspect by executing executable instructions.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any method in the first aspect or any possible implementation manner of the first aspect.

[0019] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any method in the first aspect or any possible implementation manner of the first aspect.

[0020] The present application obtains a set of alternative documents corresponding to the user's query information; analyzes each alternative document in the alternative document set, determines the subject phrases corresponding to each alternative document, and obtains a subject phrase set corresponding to the alternative document set; determines relevant information of the query information for each alternative document in the alternative document set, and obtains a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determines a scheme for a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, introduces relevant information of the query information, proposes a new concept, and proposes a scheme for determining the matching degree between the query information and the document in combination with other dimensional information, determines the matching degree between the query information and the document based on the analysis of multiple dimensional information, and achieves the technical effect of improving the matching degree between the query information and the recall result. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0022] Figure 1 A schematic diagram of the structure of a data processing system provided in one embodiment of the present application;

[0023] Figure 2a A flowchart of a data processing method provided in one embodiment of the present application;

[0024] Figure 2b A flowchart illustrating a method for determining a subject phrase and its relevance information according to an embodiment of the present application;

[0025] Figure 2c A flowchart of a marking method provided in one embodiment of the present application;

[0026] Figure 2d A scenario diagram of a data processing method provided in one embodiment of the present application;

[0027] Figure 3 A schematic diagram of the structure of a data processing device provided in one embodiment of the present application;

[0028] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present application, but should not be understood as limiting the present application.

[0030] The terms "first" and "second" in the specification, claims and drawings of the embodiments of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] First, some of the terms used in the embodiments of the present application are explained below to facilitate understanding by those skilled in the art.

[0032] LDA: A topic model that presents the topic of each document in a document set in the form of a probability distribution. It can then extract the topic (distribution) of some documents by analyzing them and perform topic clustering or text classification based on the topic (distribution).

[0033] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0034] Figure 1 A schematic diagram of a data processing system provided as an exemplary embodiment of the present application includes: a target device 11 and a server device 12; wherein:

[0035] The target device 11 is used to obtain the query information input by the user through the input box; and send the query information to the server device 12;

[0036] The server device 12 is used to receive query information and determine a corresponding alternative document set based on the user's query information; analyze each alternative document in the alternative document set, determine the subject phrases corresponding to each alternative document, and obtain a subject phrase set corresponding to the alternative document set; determine relevant information of the query information for each alternative document in the alternative document set, and obtain a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determine a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set; and send the target recall result to the target device 11 for display by the target device 11.

[0037] Optionally, the above system may also include a management device 13, and the server device 12 may also send the alternative document set, the subject phrase set, and the related information set to the management device 13, so that the management device 13 determines the target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the related information set, and displays the target recall result.

[0038] Optionally, when the server device 12 is used to determine the target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set, it is specifically used to:

[0039] For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determining the matching degree information between the query information and the candidate document, and obtaining a plurality of matching degree information corresponding to the candidate document set;

[0040] A target recall result corresponding to the query information is determined based on the candidate document set and the plurality of matching degree information.

[0041] Optionally, the server device 12 may further send the multiple pieces of matching degree information corresponding to the determined candidate document set and the candidate document set to the management device 13 for display by the management device 13 .

[0042] The target device 11 and the management device 13 may be the same device or may not be the same device, and this application does not limit this.

[0043] The execution principles and interaction processes of the various components in this system embodiment, such as the target device 11, the server device 12, and the management device 13, can be found in the descriptions of the following method embodiments.

[0044] Figure 2a A flowchart of a data processing method provided by an exemplary embodiment of the present application is provided. The method is applicable to a server device and includes at least the following steps:

[0045] S201, obtaining a candidate document set corresponding to the user's query information;

[0046] S202: Analyze each candidate document in the candidate document set, determine the subject phrases corresponding to each candidate document, and obtain a subject phrase set corresponding to the candidate document set;

[0047] S203: Determine relevant information of the query information for each candidate document in the candidate document set, and obtain a relevant information set of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, and a similarity value with the query information is greater than a preset similarity value;

[0048] S204: Determine a target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set.

[0049] Specifically, the aforementioned query information can be obtained from a target device, which can be a user device, such as a personal computer (PC), or a mobile device, such as a mobile phone, tablet computer, or other hardware device with various operating systems, touch screens, and / or displays. Optionally, the target device can also be an enterprise's control device, which is not limited in this application.

[0050] Furthermore, the method further includes: receiving query information determined by a target device based on a user operation; and determining a candidate document set using the query information. Specifically, the candidate document set is a collection of one or more candidate documents, each of which can be a document directly determined based on the number of overlaps between the query information and the word content of the document, as in the prior art.

[0051] Users can Figure 1 In the input box shown, query information is entered. After the target device obtains the query information entered by the user, it can send a query request containing the query information to the server device, so that the server device can determine the candidate document set.

[0052] Furthermore, in the aforementioned step S202, each candidate document in the candidate document set is analyzed to determine the subject phrase group corresponding to each candidate document. In the process of obtaining the subject phrase group set corresponding to the candidate document set, the candidate document is analyzed to determine the subject phrase group corresponding to each candidate document, including:

[0053] The candidate document is input into a preset LDA model to determine an analysis result corresponding to the candidate document, wherein the analysis result includes a subject phrase corresponding to the candidate document.

[0054] Specifically, the method for determining the aforementioned LDA model may include: obtaining a sample information set, and using the sample information set to train an LDA model.

[0055] In some optional embodiments of this application, see Figure 2bAs shown, the sample information set can be a sample document set. When the sample document set is used to train the LDA model, it can specifically include text cleaning, word segmentation, and stop word processing on each sample document in the sample document set to obtain a bag-of-words model corresponding to each sample document, thereby obtaining a bag-of-words model set corresponding to the sample document set, and then training the LDA model based on the bag-of-words model set; accordingly, the alternative documents are processed, including text cleaning, word segmentation, and stop word processing on the alternative documents. After processing the alternative documents, the corresponding bag-of-words model is also obtained. The bag-of-words model is used as the input parameter of the LDA model, and the LDA model is executed to obtain an analysis result, wherein the analysis result includes a subject phrase, and may also include correlation information between the subject phrase and the alternative document, wherein the correlation information may specifically include a correlation value.

[0056] Furthermore, each alternative document in the alternative document set may correspond to multiple query items. For each alternative document, when at least some of the multiple query items corresponding to it are used as the query content, the corresponding alternative document can be matched; the multiple query items corresponding to the alternative document may include the aforementioned query information, and each alternative document and its corresponding multiple query items can be stored in the corresponding correspondence information.

[0057] Optionally, one or more query items among the multiple query items corresponding to each alternative document, whose similarity with the query information is higher than the preset similarity, are the relevant information of the query information for the current alternative document; that is, the relevant information of the query information for the alternative document is the query item among the multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than the preset similarity value.

[0058] Optionally, the query information and the related information may be different, that is, the related information of the query information for the candidate document may be a query item among multiple query items corresponding to the candidate document, whose similarity value with the query information is greater than a preset similarity value but which is different from the query information.

[0059] For example: when the alternative document is a method document for hailing a taxi using map software A, its corresponding three query items may include: hailing a taxi, map A; hailing a taxi, software A; calling a taxi, map A; when the query information is: hailing a taxi, map A, hailing a taxi, software A and calling a taxi, map A are query information for relevant information on the method document for hailing a taxi using map software A.

[0060] Furthermore, in the aforementioned step S203, determining the target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set includes:

[0061] S2031. For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determine the matching degree information between the query information and the candidate document, and obtain multiple matching degree information corresponding to the candidate document set;

[0062] S2032: Determine a target recall result corresponding to the query information based on the candidate document set and the plurality of matching degree information.

[0063] Specifically, for each candidate document, using the subject phrase corresponding to the candidate document and the relevant information of the query information on the candidate document, determining the matching degree information between the query information and the candidate document includes:

[0064] If the relevant information and the subject phrase meet a first preset condition, determining the matching degree information according to the first preset condition;

[0065] If the relevant information and the subject phrase do not satisfy the first preset condition, the title information corresponding to the candidate document is obtained, and the matching degree information is determined based on the query information and the title information.

[0066] Specifically, when the relevant information is contained in a subject phrase group, that is, when it belongs to a word in the subject phrase group, the relevant information and the subject phrase group are deemed to meet the first preset condition. When the relevant information contains multiple words, when the multiple words all belong to the words in the subject phrase group, or when the ratio of the number of words in the multiple words that belong to the subject phrase group to the total number of multiple words is greater than the first preset ratio, the relevant information and the subject phrase group are deemed to meet the first preset condition. Determining the matching degree information according to the first preset condition can be specifically as follows: according to the first mapping relationship information corresponding to the first preset condition, determining the matching degree information is the first matching result in the first mapping relationship information.

[0067] Accordingly, when the relevant information is not included in the subject phrase, it is deemed that the relevant information and the subject phrase do not satisfy the first preset condition.

[0068] Optionally, when the relevant information contains multiple words, when the multiple words do not belong to the words in the subject phrase group, or the ratio of the number of words in the multiple words that belong to the subject phrase group to the total number of the multiple words is less than a second preset ratio, it can be regarded that the relevant information and the subject phrase group do not meet the first preset condition, and determining the matching information based on the query information and the title information includes: when the matching degree between the query information and the title information is greater than the preset matching degree, determining the matching information based on the preset matching degree.

[0069] Specifically, the aforementioned preset matching degree can be set by relevant personnel, and the preset matching degree can correspond to the second mapping relationship information. Determining the matching degree information based on the preset matching degree can specifically include determining that the matching degree information is the second matching result in the second mapping relationship information based on the second mapping relationship information.

[0070] Furthermore, when the matching degree between the query information and the title information is not greater than a preset matching degree, the matching degree information is determined to be a third matching result. The matching degree between the query information and the title information may be a text similarity between the query information and the title information, and the corresponding preset matching degree is a preset similarity value.

[0071] Furthermore, in some optional embodiments of the present application, the above method further includes:

[0072] Determine whether the query information and the subject phrase meet the second preset condition. If not, obtain the correlation value between the subject phrase and the alternative document. When the correlation value is in the first numerical range, trigger the step of using the subject phrase corresponding to the alternative document and the relevant information of the query information for the alternative document to determine the matching information step between the query information and the alternative document.

[0073] When the correlation value is not within the first numerical interval, it is determined whether the correlation value is within the second numerical interval. If so, matching information is determined based on the second numerical interval; wherein the second numerical interval is smaller than the first numerical interval. The second numerical interval being smaller than the first numerical interval means that the maximum value in the second numerical interval is smaller than the minimum value in the first numerical interval.

[0074] Optionally, when the query information is included in the subject phrase, the query information and the subject phrase are considered to meet the second preset condition; when the query information is not included in the subject phrase, the query information and the subject phrase do not meet the second preset condition.

[0075] Furthermore, when the query information contains multiple words, when the multiple words are all included in the subject phrase group, or the ratio of the number of words in the multiple words that belong to the subject phrase group to the total number of the multiple words is greater than the third preset ratio, the query information and the subject phrase group are deemed to meet the second preset condition; when the multiple words do not belong to the subject phrase group, or the ratio of the number of words in the multiple words that belong to the subject phrase group to the total number of the multiple words is less than the fourth preset ratio, the query information and the subject phrase group are deemed not to meet the second preset condition.

[0076] Optionally, the aforementioned first numerical interval can be [95%, 100%], and the second numerical interval can be [50%, 95%]. Determining the matching information based on the second numerical interval can specifically include: based on the third mapping relationship information corresponding to the second numerical interval, determining that the matching information is the fourth matching result in the third mapping relationship information; if the correlation value is not in the second numerical interval, determining that the matching information is the fifth matching result.

[0077] It should be noted that the first to fifth matching results involved in this application are used to indicate the final matching degree between the query information and the first recall result. Specifically, it can be a specific matching degree value, or it can be a score value used to indicate the degree of matching. This application does not limit this.

[0078] Optionally, determining the matching information based on the query information and the title information includes: when the query information and the title information meet a third preset condition, based on the third preset condition and the fourth mapping relationship information, determining that the matching information is the second matching result in the fourth mapping relationship information; if not, determining that the matching information is the third matching result.

[0079] Specifically, when the query information is included in the title information, it is considered that the query information and the title information meet the third preset condition.

[0080] Optionally, when the query information contains multiple words, when none of the multiple words belong to the words in the title information, or the ratio of the number of words in the multiple words that belong to the title information to the total number of the multiple words is less than a fifth preset ratio, it can be regarded that the query information and the title information do not meet the third preset condition; when the multiple words all belong to the words in the title information, or the ratio of the number of words in the multiple words that belong to the title information to the total number of the multiple words is greater than a sixth preset ratio, it can be regarded that the query information and the title information meet the third preset condition.

[0081] Optionally, when it is determined that the query information and the subject phrase meet the second preset condition, the matching degree information is determined to be a preset sixth matching result.

[0082] Furthermore, in the aforementioned S2032, determining a target recall result corresponding to the query information based on the candidate document set and the plurality of matching degree information includes:

[0083] Sorting the candidate documents in the candidate document set according to the plurality of matching degree information to obtain a sorting result;

[0084] The target recall result is determined according to the sorting result.

[0085] Optionally, the target recall result includes target recall documents and sorting information of the target recall documents, and the method further includes:

[0086] The target recall result is sent to the target device, so that the target device displays the target recall document according to the sorting information of the target recall document.

[0087] The higher the matching degree indicated by the corresponding matching degree information, the smaller the ranking number corresponding to the candidate document, that is, the higher the candidate document is.

[0088] Furthermore, determining the target recall result according to the ranking result may include:

[0089] Among the multiple candidate documents, a candidate document with a sorting sequence number smaller than a preset sequence number is used as a target recall document, and the sorting sequence number of the target recall document is used as the sorting information of the target recall document;

[0090] The target recall document and the ranking information of the target recall document are used as the target recall result.

[0091] The number of the target recall documents may be one or more. When the target device displays the target recall documents according to the sorting information of the target recall documents, the target device may display the target recall documents in a manner such that the smaller the sorting number, the closer the display position.

[0092] Optionally, the aforementioned target recall result may also only include the target recall document.

[0093] The above-mentioned scheme can further determine the query information and its corresponding alternative document set, as well as the matching information with each alternative document in the alternative document set based on the relevant information of the query information, and then determine the target recall result based on the matching information between the query information and each alternative document, thereby improving the matching degree and accuracy of determining the query information and its corresponding target recall result, and also optimizing the display of the target recall result set, so that the target recall result with a higher matching degree with the query information can be displayed in a higher position.

[0094] Optionally, for target recall results with the same corresponding sorting sequence number, the target devices can be displayed in any order.

[0095] Furthermore, the matching degree values or score values corresponding to the aforementioned first matching result, second matching result, third matching result, fourth matching result, fifth matching result, and sixth matching result may be set by relevant personnel.

[0096] Optionally, the first matching result, the second matching result, and the sixth matching result may be the same, and the third matching result and the fourth matching result may be the same, for example: the score value corresponding to the first matching result, the second matching result, and the score value corresponding to the sixth matching result is 2, the score value corresponding to the third matching result and the score value corresponding to the fourth matching result is 1, and the score value corresponding to the fifth matching result is 0.

[0097] Furthermore, the aforementioned matching degree information may also be used to mark the score values of the query information and the corresponding candidate documents, and the score value is used to indicate the matching degree of the query information and the candidate documents.

[0098] Furthermore, the above method further includes: sending the plurality of matching degree information corresponding to the candidate document set and the candidate document set to a management device for analysis by relevant personnel.

[0099] The above method can be applied to searching for documents, and can be specifically applied to search engines or applications with search functions, which is not limited in this application.

[0100] The present application obtains a set of alternative documents corresponding to the user's query information; analyzes each alternative document in the alternative document set, determines the subject phrases corresponding to each alternative document, and obtains a subject phrase set corresponding to the alternative document set; determines the relevant information corresponding to each alternative document in the alternative document set for the query information, and obtains a relevant information set corresponding to the alternative document set, wherein the relevant information is a query item whose similarity value with the query information is greater than a preset similarity value among multiple query items corresponding to the corresponding alternative documents; determines a scheme for a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, introduces relevant information of the query information, proposes a new concept, and proposes a scheme for determining the matching degree between the query information and the document in combination with other dimensional information. Based on the analysis of multiple dimensional information, the matching degree between the query information and the document is determined, thereby achieving a technical effect of improving the matching degree between the query information and the recall result.

[0101] The following is a further explanation of the solution of this application in combination with specific scenarios:

[0102] Scenario 1:

[0103] This solution can be used to automatically mark the corresponding score values according to the different correlations between the query information and the subject set (corresponding to the aforementioned subject phrases), the related information of the query information and the subject set, and the candidate documents corresponding to the subject set and the query information. For details, please refer to Figure 2cAs shown, when the query information is a subset of the subject set, the score value is marked as 2. When the query information does not belong to the subset of the subject set, it is determined whether the relevance value between the subject set and the document (corresponding to the aforementioned alternative document) reaches more than 95%. If so, it is determined whether the same target word of the query information (corresponding to the relevant information of the aforementioned query information) belongs to the subset of the subject set. If so, the score value is marked as 2. If not, it is determined whether the query information is included in the title of the document. If so, the score value is marked as 2. If not, the score value is marked as 1. Further, if the relevance value between the subject set and the document does not reach more than 95%, it is determined whether the relevance value between the subject set and the document reaches more than 50%. If so, the score value is marked as 1. If not, the score value is marked as 0.

[0104] Scenario 2

[0105] See also Figure 2d As shown, after receiving the query information, the server device directly determines the candidate document set based on the number of overlaps between the query information and the words in the document content. The original order of the multiple candidate documents in the candidate document set is: candidate document A, candidate document B, candidate document C, candidate document D, candidate document E. After determining the matching degree information corresponding to the candidate documents through this solution, if the scores determined in sequence are 1, 2, 0, 2, and 1, the candidate documents in the candidate document set are sorted from large to small according to the corresponding scores, obtaining the ranking information corresponding to the candidate document set. The candidate document set and the ranking information are then sent to the target device, causing the target device to display the candidate document set (i.e., the target recall document) according to the ranking information. Specifically, the display order of the multiple candidate documents in the candidate document set displayed by the target device is: candidate document B, candidate document D, candidate document A, candidate document E, and candidate document C. The ranking information is related to the display positions of the multiple candidate documents in the target candidate document set.

[0106] An exemplary embodiment of the present application further provides a data processing method, which may include:

[0107] Determine corresponding query information according to the user's operation instruction;

[0108] Sending a query request corresponding to the query information to a server device, so that the server device obtains a set of alternative documents corresponding to the user's query information according to the query request; analyzing each alternative document in the alternative document set, determining a subject phrase corresponding to each alternative document, and obtaining a subject phrase set corresponding to the alternative document set; determining relevant information of the query information for each alternative document in the alternative document set, and obtaining a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determining a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, and feeding back the target recall result;

[0109] Receive and display the target recall result.

[0110] Specifically, the specific implementation methods corresponding to this embodiment, for example, the method for the server device to determine the subject phrase set corresponding to the recall result set, and the method for determining the matching degree information set of the recall result set can be referred to the above content and will not be repeated here.

[0111] Figure 3 A schematic structural diagram of a data processing device provided as an exemplary embodiment of the present application;

[0112] The device includes: an acquisition module 31, an analysis module 32, a first determination module 33, and a second determination module 34; wherein:

[0113] An acquisition module 31 is used to acquire a candidate document set corresponding to the user's query information;

[0114] An analysis module 32 is configured to analyze each candidate document in the candidate document set, determine a subject phrase group corresponding to each candidate document, and obtain a subject phrase group group corresponding to the candidate document set;

[0115] A first determining module 33 is configured to determine relevant information of the query information for each candidate document in the candidate document set, and obtain a set of relevant information of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, the similarity value of which to the query information is greater than a preset similarity value;

[0116] The second determining module 34 is configured to determine a target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set.

[0117] Optionally, when the aforementioned apparatus is used to determine the target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set, it is specifically used to:

[0118] For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determining the matching degree information between the query information and the candidate document, and obtaining a plurality of matching degree information corresponding to the candidate document set;

[0119] A target recall result corresponding to the query information is determined based on the candidate document set and the plurality of matching degree information.

[0120] Optionally, when the aforementioned device is used to analyze each candidate document and determine the subject phrase group corresponding to the candidate document, it is specifically used to:

[0121] The candidate document is input into a preset LDA model to determine an analysis result corresponding to the candidate document, wherein the analysis result includes a subject phrase corresponding to the candidate document.

[0122] Optionally, when the aforementioned apparatus is used to determine, for each candidate document, the matching information between the query information and the candidate document using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document, the apparatus is specifically used to:

[0123] If the relevant information and the subject phrase meet a first preset condition, determining the matching degree information according to the first preset condition;

[0124] If the relevant information and the subject phrase do not satisfy the first preset condition, the title information corresponding to the candidate document is obtained, and the matching degree information is determined based on the query information and the title information.

[0125] Optionally, when the aforementioned apparatus is used to determine the matching information based on the query information and the title information, it is specifically used to:

[0126] When the matching degree between the query information and the title information is greater than a preset matching degree, the matching degree information is determined based on the preset matching degree.

[0127] Optionally, the analysis result further includes a relevance value between the subject phrase and the candidate document, and the device is further configured to:

[0128] Determine whether the query information and the subject phrase meet the second preset condition, if not,

[0129] Obtain the relevance value between the subject phrase and the candidate document. When the relevance value is in a first numerical range, trigger the step of using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document to determine the matching information between the query information and the candidate document.

[0130] Optionally, the aforementioned device is further used for:

[0131] When the correlation value is not in the first numerical interval, it is determined whether the correlation value is in a second numerical interval. If so, the matching information is determined according to the second numerical interval; wherein the second numerical interval is smaller than the first numerical interval.

[0132] Optionally, when the aforementioned apparatus is used to determine the target recall result corresponding to the query information based on the candidate document set and the plurality of matching degree information, it is specifically used to:

[0133] Sorting the candidate documents in the candidate document set according to the plurality of matching degree information to obtain a sorting result;

[0134] The target recall result is determined according to the sorting result.

[0135] Optionally, the target recall result includes the target recall document and the sorting information of the target recall document, and the aforementioned device is further used to:

[0136] The target recall result is sent to a target device, so that the target device displays the target recall document according to the sorting information of the target recall document.

[0137] Furthermore, the present application also provides a device, comprising:

[0138] A determination module, configured to determine corresponding query information according to a user's operation instruction;

[0139] A sending module, configured to send a query request corresponding to the query information to a server device, so that the server device obtains a set of alternative documents corresponding to the user's query information according to the query request; analyzes each alternative document in the alternative document set, determines a subject phrase corresponding to each alternative document, and obtains a subject phrase set corresponding to the alternative document set; determines relevant information of the query information for each alternative document in the alternative document set, and obtains a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determines a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, and feeds back the target recall result;

[0140] The display module is used to receive and display the target recall result.

[0141] It should be understood that the device embodiments and the method embodiments may correspond to each other, and similar descriptions may refer to the method embodiments. To avoid repetition, they will not be described in detail here. Specifically, the device can perform the above-mentioned method embodiments, and the aforementioned and other operations and / or functions of each module in the device are the corresponding processes in each method in the above-mentioned method embodiments, which will not be described in detail here for the sake of brevity.

[0142] The apparatus of the embodiment of the present application is described above from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, or can be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps in the above method embodiment in conjunction with its hardware.

[0143] Figure 4 : is a schematic block diagram of an electronic device provided in an embodiment of the present application, and the electronic device may include:

[0144] The memory 401 and the processor 402 are configured to store computer programs and transmit the program code to the processor 402. In other words, the processor 402 can call and run the computer program from the memory 401 to implement the method in the embodiment of the present application.

[0145] For example, the processor 402 may be configured to execute the above method embodiments according to instructions in the computer program.

[0146] In some embodiments of the present application, the processor 402 may include but is not limited to:

[0147] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0148] In some embodiments of the present application, the memory 401 includes but is not limited to:

[0149] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0150] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory 401 and executed by the processor 402 to implement the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0151] like Figure 4 As shown, the electronic device may further include:

[0152] The transceiver 403 may be connected to the processor 402 or the memory 401 .

[0153] The processor 402 may control the transceiver 403 to communicate with other devices. Specifically, the processor 402 may send information or data to other devices or receive information or data sent by other devices. The transceiver 403 may include a transmitter and a receiver. The transceiver 403 may further include one or more antennas.

[0154] It should be understood that the various components in the electronic device are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0155] The present application also provides a computer storage medium having a computer program stored thereon, which, when executed by a computer, enables the computer to perform the method of the above-mentioned method embodiment. In other words, the present application also provides a computer program product containing instructions, which, when executed by a computer, enables the computer to perform the method of the above-mentioned method embodiment.

[0156] When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0157] According to one or more embodiments of the present application, a data processing method is provided, including:

[0158] Obtain the candidate document set corresponding to the user's query information;

[0159] Analyzing each candidate document in the candidate document set, determining a subject phrase group corresponding to each candidate document, and obtaining a subject phrase group group corresponding to the candidate document set;

[0160] Determining relevant information of the query information for each candidate document in the candidate document set, and obtaining a relevant information set of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, and a similarity value with the query information is greater than a preset similarity value;

[0161] A target recall result corresponding to the query information is determined based on the candidate document set, the subject phrase set, and the related information set.

[0162] According to one or more embodiments of the present application, determining a target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set includes:

[0163] For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determining the matching degree information between the query information and the candidate document, and obtaining a plurality of matching degree information corresponding to the candidate document set;

[0164] A target recall result corresponding to the query information is determined based on the candidate document set and the plurality of matching degree information.

[0165] According to one or more embodiments of the present application, for each candidate document, analyzing the candidate document to determine the subject phrase corresponding to the candidate document includes:

[0166] The candidate document is input into a preset LDA model to determine an analysis result corresponding to the candidate document, wherein the analysis result includes a subject phrase corresponding to the candidate document.

[0167] According to one or more embodiments of the present application, for each candidate document, determining, using a subject phrase corresponding to the candidate document and relevant information of the query information related to the candidate document, a degree of matching between the query information and the candidate document includes:

[0168] If the relevant information and the subject phrase meet a first preset condition, determining the matching degree information according to the first preset condition;

[0169] If the relevant information and the subject phrase do not satisfy the first preset condition, the title information corresponding to the candidate document is obtained, and the matching degree information is determined based on the query information and the title information.

[0170] According to one or more embodiments of the present application, determining the matching information based on the query information and the title information includes:

[0171] When the matching degree between the query information and the title information is greater than a preset matching degree, the matching degree information is determined based on the preset matching degree.

[0172] According to one or more embodiments of the present application, the analysis result further includes a relevance value between the subject phrase and the candidate document, and the method further includes:

[0173] Determine whether the query information and the subject phrase meet the second preset condition, if not,

[0174] Obtain the relevance value between the subject phrase and the candidate document. When the relevance value is in a first numerical range, trigger the step of using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document to determine the matching information between the query information and the candidate document.

[0175] According to one or more embodiments of the present application, the method further includes:

[0176] When the correlation value is not in the first numerical interval, it is determined whether the correlation value is in a second numerical interval. If so, the matching information is determined according to the second numerical interval; wherein the second numerical interval is smaller than the first numerical interval.

[0177] According to one or more embodiments of the present application, determining a target recall result corresponding to the query information based on the candidate document set and the plurality of matching degree information includes:

[0178] Sorting the candidate documents in the candidate document set according to the plurality of matching degree information to obtain a sorting result;

[0179] The target recall result is determined according to the sorting result.

[0180] According to one or more embodiments of the present application, the target recall result includes the target recall document and ranking information of the target recall document, and the method further includes:

[0181] The target recall result is sent to a target device, so that the target device displays the target recall document according to the sorting information of the target recall document.

[0182] According to one or more embodiments of the present application, a data processing device is provided, comprising: an acquisition module, an analysis module, a first determination module, and a second determination module; wherein:

[0183] The acquisition module is used to obtain the candidate document set corresponding to the user's query information;

[0184] An analysis module, configured to analyze each candidate document in the candidate document set, determine a subject phrase group corresponding to each candidate document, and obtain a subject phrase group group corresponding to the candidate document set;

[0185] a first determining module configured to determine relevant information of the query information for each candidate document in the candidate document set, and obtain a set of relevant information of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, the query items having a similarity value with the query information greater than a preset similarity value;

[0186] The second determining module is configured to determine a target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set.

[0187] According to one or more embodiments of the present application, when the aforementioned apparatus is used to determine the target recall result corresponding to the query information based on the candidate document set, the subject phrase set, and the related information set, it is specifically used to:

[0188] For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determining the matching degree information between the query information and the candidate document, and obtaining a plurality of matching degree information corresponding to the candidate document set;

[0189] A target recall result corresponding to the query information is determined based on the candidate document set and the plurality of matching degree information.

[0190] According to one or more embodiments of the present application, when the aforementioned apparatus is used to analyze each candidate document and determine the subject phrase group corresponding to the candidate document, it is specifically used to:

[0191] The candidate document is input into a preset LDA model to determine an analysis result corresponding to the candidate document, wherein the analysis result includes a subject phrase corresponding to the candidate document.

[0192] According to one or more embodiments of the present application, when the aforementioned apparatus is used to determine, for each candidate document, the matching information between the query information and the candidate document using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document, the apparatus is specifically configured to:

[0193] If the relevant information and the subject phrase meet a first preset condition, determining the matching degree information according to the first preset condition;

[0194] If the relevant information and the subject phrase do not satisfy the first preset condition, the title information corresponding to the candidate document is obtained, and the matching degree information is determined based on the query information and the title information.

[0195] According to one or more embodiments of the present application, when the aforementioned apparatus is used to determine the matching information based on the query information and the title information, it is specifically used to:

[0196] When the matching degree between the query information and the title information is greater than a preset matching degree, the matching degree information is determined based on the preset matching degree.

[0197] According to one or more embodiments of the present application, the analysis result further includes a relevance value between the subject phrase and the candidate document, and the apparatus is further configured to:

[0198] Determine whether the query information and the subject phrase meet the second preset condition, if not,

[0199] Obtain the relevance value between the subject phrase and the candidate document. When the relevance value is in a first numerical range, trigger the step of using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document to determine the matching information between the query information and the candidate document.

[0200] According to one or more embodiments of the present application, the aforementioned device is further used to:

[0201] When the correlation value is not in the first numerical interval, it is determined whether the correlation value is in a second numerical interval. If so, the matching information is determined according to the second numerical interval; wherein the second numerical interval is smaller than the first numerical interval.

[0202] According to one or more embodiments of the present application, when the aforementioned apparatus is used to determine the target recall result corresponding to the query information based on the candidate document set and the multiple matching degree information, it is specifically used to:

[0203] Sorting the candidate documents in the candidate document set according to the plurality of matching degree information to obtain a sorting result;

[0204] The target recall result is determined according to the sorting result.

[0205] According to one or more embodiments of the present application, the target recall result includes the target recall document and the sorting information of the target recall document. The aforementioned apparatus is further configured to:

[0206] The target recall result is sent to a target device, so that the target device displays the target recall document according to the sorting information of the target recall document.

[0207] According to one or more embodiments of the present application, a device is provided, including:

[0208] A determination module, configured to determine corresponding query information according to a user's operation instruction;

[0209] A sending module, configured to send a query request corresponding to the query information to a server device, so that the server device obtains a set of alternative documents corresponding to the user's query information according to the query request; analyzes each alternative document in the alternative document set, determines a subject phrase corresponding to each alternative document, and obtains a subject phrase set corresponding to the alternative document set; determines relevant information of the query information for each alternative document in the alternative document set, and obtains a relevant information set of the query information for the alternative document set, wherein the relevant information is a query item among multiple query items corresponding to the alternative document, whose similarity value with the query information is greater than a preset similarity value; determines a target recall result corresponding to the query information based on the alternative document set, the subject phrase set, and the relevant information set, and feeds back the target recall result;

[0210] The display module is used to receive and display the target recall result.

[0211] According to one or more embodiments of the present application, an electronic device is provided, including:

[0212] processor; and

[0213] a memory for storing executable instructions for the processor;

[0214] The processor is configured to execute the aforementioned data processing methods by executing executable instructions.

[0215] According to one or more embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the aforementioned data processing methods are implemented.

[0216] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0217] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0218] Modules described as separate components may or may not be physically separate, and components displayed as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected based on actual needs to achieve the purpose of the present embodiment. For example, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module.

[0219] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A data processing method, characterized in that: include: Obtain the candidate document set corresponding to the user's query information; Analyzing each candidate document in the candidate document set, determining a subject phrase group corresponding to each candidate document, and obtaining a subject phrase group group corresponding to the candidate document set; Determining relevant information of the query information for each candidate document in the candidate document set, and obtaining a relevant information set of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, and a similarity value with the query information is greater than a preset similarity value; For each candidate document in the candidate document set, using the subject phrase corresponding to the candidate document and the relevant information corresponding to the candidate document in the query information, determining the matching degree information between the query information and the candidate document, and obtaining a plurality of matching degree information corresponding to the candidate document set; A target recall result corresponding to the query information is determined based on the candidate document set and the plurality of matching degree information.

2. The method according to claim 1, characterized in that For each candidate document, the candidate document is analyzed to determine the subject phrases corresponding to the candidate document, including: The candidate document is input into a preset LDA model to determine an analysis result corresponding to the candidate document, wherein the analysis result includes a subject phrase corresponding to the candidate document.

3. The method according to claim 2, characterized in that For each candidate document, using the subject phrase corresponding to the candidate document and the relevant information of the query information on the candidate document, determining the matching degree information between the query information and the candidate document includes: If the relevant information and the subject phrase meet a first preset condition, determining the matching degree information according to the first preset condition; If the relevant information and the subject phrase do not satisfy the first preset condition, the title information corresponding to the candidate document is obtained, and the matching degree information is determined based on the query information and the title information.

4. The method according to claim 3, characterized in that Determining the matching degree information based on the query information and the title information includes: When the matching degree between the query information and the title information is greater than a preset matching degree, the matching degree information is determined based on the preset matching degree.

5. The method according to claim 4, characterized in that The analysis result also includes a relevance value between the subject phrase and the candidate document. The method further includes: Determine whether the query information and the subject phrase meet the second preset condition, if not, Obtain the relevance value between the subject phrase and the candidate document. When the relevance value is in a first numerical range, trigger the step of using the subject phrase corresponding to the candidate document and the relevant information of the query information for the candidate document to determine the matching information between the query information and the candidate document.

6. The method according to claim 5, characterized in that The method further comprises: When the correlation value is not in the first numerical interval, it is determined whether the correlation value is in a second numerical interval. If so, the matching information is determined according to the second numerical interval; wherein the second numerical interval is smaller than the first numerical interval.

7. The method according to claim 1, characterized in that Determining a target recall result corresponding to the query information based on the candidate document set and the plurality of matching degree information includes: Sorting the candidate documents in the candidate document set according to the plurality of matching degree information to obtain a sorting result; The target recall result is determined according to the sorting result.

8. The method according to claim 7, characterized in that The target recall result includes the target recall document and the sorting information of the target recall document. The method further includes: The target recall result is sent to a target device, so that the target device displays the target recall document according to the sorting information of the target recall document.

9. A data processing device, characterized in that: include: an acquisition module, an analysis module, a first determination module, and a second determination module; wherein: The acquisition module is used to obtain the candidate document set corresponding to the user's query information; An analysis module, configured to analyze each candidate document in the candidate document set, determine a subject phrase group corresponding to each candidate document, and obtain a subject phrase group group corresponding to the candidate document set; a first determining module configured to determine relevant information of the query information for each candidate document in the candidate document set, and obtain a set of relevant information of the query information for the candidate document set, wherein the relevant information is a query item among multiple query items corresponding to the candidate documents, the query items having a similarity value with the query information greater than a preset similarity value; The second determination module is used to determine the matching information between the query information and the alternative document for each alternative document in the alternative document set by using the subject phrase corresponding to the alternative document and the relevant information corresponding to the alternative document of the query information, and obtain multiple matching information corresponding to the alternative document set; and determine the target recall result corresponding to the query information based on the alternative document set and the multiple matching information.

10. An electronic device, characterized in that: include: processor; as well as a memory for storing executable instructions of the processor; The processor is configured to perform the method according to any one of claims 1 to 8 by executing the executable instructions.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images