Search result recall method, device and readable storage medium

Abstract

The present invention provides a search result recall method, device and readable storage medium, the search result recall method comprising: obtaining user input characters; segmenting the user input characters to obtain user input words; searching based on the user input words to obtain a first data source; based on the number of the first data source being less than a first threshold, obtaining hot search tags corresponding to the user input words, performing a secondary search based on the hot search tags to obtain a second data source, and displaying the first data source and the second data source. In the technical solution of the present invention, by setting a first threshold, in the case where the search results recalled based on the user input words (keywords) are too few, the hot search tags corresponding to the user input words are obtained, and a secondary search is performed based on the hot search tags, thereby avoiding the problem that the user needs to repeatedly try different keywords to search when the search results recalled based on keywords are too few, and quickly finding the target result.

Description

Technical Field

[0001] The present invention relates to the field of search technology, and in particular to a search result recall method, device and readable storage medium. Background Art

[0002] In the related art, the search technology has the following problems in processing search results. First, when there are too few search results based on keyword recall, users need to repeatedly try different keywords to find the desired results. Second, in the search result recall, it is difficult to consider the search results of the larger extension-related and time-related vocabulary of the search term. Summary of the invention

[0003] The present invention aims to solve at least one of the above technical problems.

[0004] To this end, a first object of the present invention is to provide a search result recall method.

[0005] A second object of the present invention is to provide a search result recall device.

[0006] A third object of the present invention is to provide a readable storage medium.

[0007] To achieve the first purpose of the present invention, the technical solution of the present invention provides a search result recall method including: obtaining user input characters; segmenting the user input characters to obtain user input vocabulary; searching according to the user input vocabulary to obtain a first data source; based on the number of the first data source being less than a first threshold, obtaining hot search tags corresponding to the user input vocabulary, performing a secondary search according to the hot search tags to obtain a second data source, and displaying the first data source and the second data source.

[0008] In this technical solution, by setting a first threshold, when there are too few search results based on the user input vocabulary (keywords), the hot search tags corresponding to the user input vocabulary are obtained, and a secondary search is performed based on the hot search tags, thereby avoiding the problem that the user needs to repeatedly try different keywords to search when there are too few search results based on keyword recall. The problem of insufficient search results is solved from a semantic level, and the number of user searches is effectively reduced, and the target results are quickly found, thereby improving the user experience. In this technical solution, the first data source and the second data source are displayed simultaneously, and the search results are displayed more intuitively, effectively covering the user search content, thereby improving the user experience.

[0009] In addition, the technical solution provided by the present invention may also have the following additional technical features:

[0010] In the above technical solution, obtaining the hot search tag corresponding to the user input word specifically includes: obtaining the hot search tag corresponding to the user input word according to the hot search tag map.

[0011] In the search result recall, the relevance between information and search terms is the core, and it is difficult to consider the search results of the larger extension-related and time-related vocabulary of the search terms. In this technical solution, the hot search tags corresponding to the user input vocabulary are obtained through the hot search tag map. The hot search tags in the hot search tag map are time-sensitive, and the timeliness problem can be solved from the search input link. In this technical solution, the hot search tags are obtained through the hot search tag map, and the hot search tags are used for secondary retrieval, which increases the simultaneous and relevant recall results, improves the recall quality of the search, and solves the problems of too little search recall and poor timeliness of semantic results.

[0012] In any of the above technical solutions, before executing to obtain the user input characters, it also includes: collecting hot search tags of the search object, which may be one or more hot search tags; setting a hot search tag group, which includes all hot search tags; and constructing a hot search tag map based on the hot search tag group.

[0013] In the technical scheme, first, according to the search object, the hot search tag of the search object is obtained, and then, according to the hot search tag, a hot search tag group is set, and finally, a hot search tag map is obtained. For example, the search object can be the same product or the same category of products, and the hot search tags of the same product or the same category of products are obtained, and the hot search tags are set as hot search tag groups to obtain a hot search tag map. During the search execution process, the number of recall results is confirmed by a first threshold. If the number is less than the set first threshold, the hot search tag is obtained through the hot search tag map, and the hot search tag is automatically called to perform a secondary search, and the recalled results are included in all the recalled results for sorting, which solves the problems of too few search recalls and poor timeliness of semantic results, and improves the user experience.

[0014] In any of the above technical solutions, the search objects include product names, industry names and / or identity identification numbers.

[0015] In this technical solution, the search object can be information such as product name, industry name and / or identity identification number (ID identification). By setting the search object to product name, industry name and / or identity identification number, the scope of user search keywords can be effectively expanded, so that the search recall results are more in line with user needs and the user experience is improved.

[0016] In any of the above technical solutions, constructing a hot search tag map according to the hot search tag group specifically includes: based on the hot search tag group including a hot search tag, setting the hot search tag as the hot search tag map.

[0017] When there is only one hot search tag in the hot search tag group, there is no need to calculate the co-occurrence probability of the hot search tags. It is only necessary to set one hot search tag as a hot search tag map. This technical solution provides a method for constructing a hot search tag map when there is only one hot search tag.

[0018] In any of the above technical solutions, a hot search tag map is constructed according to the hot search tag group, specifically including: based on the fact that the hot search tag group includes multiple hot search tags, the co-occurrence probability of the hot search tag group is calculated in the form of word vectors to obtain the correlation coefficient of any two hot search tags, and the two hot search tags with a correlation coefficient greater than a second threshold are set to be correlated to form a hot search tag map.

[0019] In search result hits, relevance is an important factor. In this technical solution, when a hot search tag group includes multiple hot search tags, the co-occurrence probability is calculated for the hot search tag group in the form of word vectors to obtain the correlation coefficient. By setting a second threshold, it is possible to screen out whether two hot search tags are related, and then obtain a hot search tag map. By adopting the form of a hot search tag map, the timeliness and relevance of the search input are ensured. In this technical solution, after the hot search tags are mapped, they are applied to the field of search result recall, which improves the quality of the search recall results. At the same time, the hot search tags enhance the timeliness of the search results.

[0020] In any of the above technical solutions, displaying the first data source and the second data source specifically includes: sorting the first data source and the second data source according to the sorting weights and then displaying them.

[0021] In this technical solution, the first data source and the second data source are sorted according to the sorting weights and then displayed to the user. By setting the sorting weights, the search recall results can be displayed better and more intuitively, so that the user can first see the search recall results with higher relevance, thereby improving the user experience.

[0022] In any of the above technical solutions, the sorting weight is set according to the correlation coefficient of the hot search tags.

[0023] In this technical solution, the sorting weight can be set according to the correlation coefficient of the hot search tag, and the sorting weight of the hot search tag with high correlation with the user input vocabulary can be increased, so that the user can first see the search recall results with higher correlation, thereby improving the user experience.

[0024] To achieve the second purpose of the present invention, the technical solution of the present invention provides a search result recall device, including: a memory and a processor, the memory stores a program or instruction, and the processor executes the program or instruction; wherein, when the processor executes the program or instruction, it implements the steps of the search result recall method of any technical solution of the present invention.

[0025] The search result recall device provided by the present technical solution implements the steps of the search result recall method of any technical solution of the present invention, and thus has all the beneficial effects of the search result recall method of any technical solution of the present invention, which will not be repeated here.

[0026] To achieve the third purpose of the present invention, the technical solution of the present invention provides a readable storage medium, which stores a program or instruction. When the program or instruction is executed, the steps of the search result recall method of any of the above technical solutions are implemented.

[0027] The readable storage medium provided by this technical solution implements the steps of the search result recall method of any technical solution of the present invention, and thus has all the beneficial effects of the search result recall method of any technical solution of the present invention, which will not be repeated here.

[0028] Additional aspects and advantages of the present invention will become apparent from the following description or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0030] Figure 1 This is a flowchart of a search result recall method according to an embodiment of the present invention;

[0031] Figure 2 This is a second flowchart of a search result recall method according to an embodiment of the present invention;

[0032] Figure 3 This is a third flowchart of a search result recall method according to an embodiment of the present invention;

[0033] Figure 4 This is a fourth flowchart of a search result recall method according to an embodiment of the present invention;

[0034] Figure 5 This is a fifth flowchart of a search result recall method according to an embodiment of the present invention;

[0035] Figure 6 This is a sixth flowchart of a search result recall method according to an embodiment of the present invention;

[0036] Figure 7 A schematic diagram of the composition of a search result recall device according to an embodiment of the present invention;

[0037] Figure 8 This is a seventh flowchart of a search result recall method according to an embodiment of the present invention;

[0038] Fig. 9This is an eighth flowchart of a search result recall method according to an embodiment of the present invention;

[0039] Fig.10 A schematic diagram of a milk hot search tag map according to an embodiment of the present invention;

[0040] Fig.11 This is a ninth flowchart of a search result recall method according to an embodiment of the present invention;

[0041] Fig.12 A schematic diagram of a hot search phrase according to an embodiment of the present invention;

[0042] Fig.13 A schematic diagram of the correlation coefficient of querying any two words according to an embodiment of the present invention;

[0043] Fig.14 A schematic diagram of the most relevant vocabulary group for querying any word according to an embodiment of the present invention.

[0044] in, Figure 7 The corresponding relationship between the reference numerals and the component names is as follows:

[0045] 200: search result recall device, 210: memory, 220: processor. DETAILED DESCRIPTION

[0046] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0047] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0048] Refer to the following Figures 1 to 14 The search result recall method, search result recall device 200 and readable storage medium according to some embodiments of the present invention are described.

[0049] Embodiment 1:

[0050] like Figure 1 As shown, this embodiment provides a search result recall method, comprising the following steps:

[0051] Step S102, obtaining characters input by the user;

[0052] Step S104, segmenting the user input characters to obtain the user input vocabulary;

[0053] Step S106, searching according to the words input by the user to obtain the first data source;

[0054] Step S108, based on the number of first data sources being less than a first threshold, obtaining a hot search tag corresponding to the user input word, performing a secondary search based on the hot search tag, obtaining a second data source, and displaying the first data source and the second data source.

[0055] In this embodiment, by setting a first threshold, when there are too few search results based on the user-input vocabulary (keywords) recalled, the hot search tags corresponding to the user-input vocabulary are obtained, and a secondary search is performed based on the hot search tags. This avoids the problem that when there are too few search results based on keyword recall, the user needs to repeatedly try different keywords to search. This solves the problem of insufficient search results from a semantic level, effectively reduces the number of user searches, quickly finds the target results, and improves the user experience.

[0056] In this embodiment, the first data source and the second data source are displayed simultaneously, and the search results are displayed more intuitively, effectively covering the user's search content, thereby improving the user experience.

[0057] Embodiment 2:

[0058] like Figure 2 As shown, in addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0059] Obtaining the hot search tags corresponding to the user input words includes the following steps:

[0060] Step S202: Obtain the hot search tag corresponding to the user input word according to the hot search tag map.

[0061] In the search result recall, the relevance between information and search terms is the core, and it is difficult to consider the search results of the larger extension-related and time-related words of the search terms. In this embodiment, the hot search tags corresponding to the user input words are obtained through the hot search tag map. The hot search tags in the hot search tag map are time-sensitive, and the timeliness problem can be solved from the search input link.

[0062] In this embodiment, hot search tags are obtained through a hot search tag map, and hot search tags are used for secondary retrieval, which increases simultaneous and relevant recall results, improves the recall quality of the search, and solves the problems of too little search recall and poor timeliness of semantic results.

[0063] Embodiment 3:

[0064] like Figure 3 As shown, in addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0065] Before executing to obtain the user input characters, the following steps are also included:

[0066] Step S302, collecting hot search tags of the search object, where the hot search tags can be one or more;

[0067] Step S304, setting a hot search tag group, the hot search tag group includes all hot search tags;

[0068] Step S306, constructing a hot search tag graph according to the hot search tag group.

[0069] In this embodiment, first, according to the search object, the hot search tag of the search object is obtained, and then, according to the hot search tag, a hot search tag group is set, and finally, a hot search tag map is obtained. For example, the search object can be the same product or the same category of products, and the hot search tags of the same product or the same category of products are obtained. The hot search tags are set as hot search tag groups, and then a hot search tag map is obtained. During the search execution process, the number of recall results is confirmed through a first threshold. If the number is less than the set first threshold, the hot search tag is obtained through the hot search tag map, and the hot search tag is automatically called to perform a secondary search, and the recalled results are included in all the recalled results for sorting, which solves the problems of too few search recalls and poor timeliness of semantic results, and improves user experience.

[0070] Embodiment 4:

[0071] In addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0072] Search objects include product names, industry names and / or identification numbers.

[0073] In this embodiment, the search object can be information such as product name, industry name and / or identity identification number (ID identification). By setting the search object to product name, industry name and / or identity identification number, the scope of user search keywords can be effectively expanded, so that the search recall results are more in line with user needs and the user experience is improved.

[0074] Embodiment 5:

[0075] like Figure 4 As shown, in addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0076] According to the hot search tag group, a hot search tag map is constructed, which specifically includes the following steps:

[0077] Step S402: based on the hot search tag group including a hot search tag, set the hot search tag as a hot search tag map.

[0078] When the hot search tag group has only one hot search tag, there is no need to calculate the co-occurrence probability of the hot search tags. It is only necessary to set one hot search tag as a hot search tag map. This embodiment provides a method for constructing a hot search tag map when there is only one hot search tag.

[0079] Embodiment 6:

[0080] like Figure 5 As shown, in addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0081] According to the hot search tag group, a hot search tag map is constructed, which specifically includes the following steps:

[0082] Step S502, based on the fact that the hot search tag group includes multiple hot search tags, the co-occurrence probability of the hot search tag group is calculated in the form of word vectors to obtain the correlation coefficient of any two hot search tags, and the two hot search tags with correlation coefficients greater than the second threshold are set to be correlated to form a hot search tag map.

[0083] In search result hits, relevance is an important factor. In this embodiment, when a hot search tag group includes multiple hot search tags, the co-occurrence probability is calculated for the hot search tag group in a word vector manner to obtain a correlation coefficient. By setting a second threshold, it is possible to screen out whether two hot search tags are related, and then obtain a hot search tag map. By adopting the form of a hot search tag map, the timeliness and relevance of the search input are ensured.

[0084] In this embodiment, the hot search tags are mapped and applied to the search result recall field, which improves the quality of the search recall results. At the same time, the hot search tags enhance the timeliness of the search results.

[0085] Embodiment 7:

[0086] like Figure 6 As shown, in addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0087] Displaying the first data source and the second data source specifically includes the following steps:

[0088] Step S602: sort the first data source and the second data source according to the sorting weights and then display them.

[0089] In this embodiment, the first data source and the second data source are sorted according to the sorting weights and then displayed to the user. By setting the sorting weights, the search recall results can be displayed better and more intuitively, so that the user can first see the search recall results with higher relevance, thereby improving the user experience.

[0090] Embodiment 8:

[0091] In addition to the technical features of the above embodiment, this embodiment further includes the following technical features:

[0092] The sorting weight is set according to the correlation coefficient of the hot search tags.

[0093] In this embodiment, the sorting weight can be set according to the correlation coefficient of the hot search tag, and the sorting weight of the hot search tag with high correlation with the user input vocabulary can be increased, so that the user can first see the search recall results with higher correlation, thereby improving the user experience.

[0094] Embodiment 9:

[0095] like Figure 7 As shown, this embodiment provides a search result recall device 200, including: a memory 210 and a processor 220, the memory 210 stores a program or instruction, and the processor 220 executes the program or instruction; wherein, when the processor 220 executes the program or instruction, it implements the steps of the search result recall method as any embodiment of the present invention.

[0096] Embodiment 10:

[0097] This embodiment provides a readable storage medium, which stores a program or instruction. When the program or instruction is executed by the processor 220, the steps of the search result recall method of any of the above embodiments are implemented.

[0098] Embodiment 11:

[0099] In the related technologies, search technology often faces the following difficulties in processing search results: first, when there are too few search results based on keyword recall, users need to repeatedly try different keywords to find the desired results; second, in the search result recall, the relevance between information and search terms is the core, and it is difficult to consider the search results of the larger extension relevance of the search terms and timeliness related words.

[0100] In this embodiment, first, a related vocabulary label map (that is, a hot search label map) is constructed based on the hot search words (that is, hot search tags) of the same product or the same category of products. Then, during the search execution process, the number of recall results is confirmed by a first threshold. If the number is less than the set first threshold, the related hot search tags (the second threshold of the hot search tag correlation coefficient is set) are automatically called to perform a secondary search, and the recalled results are included in the sorting of all recalled results, which solves the problem of too few search recalls and poor timeliness of semantic results.

[0101] This embodiment is a search result recall method, that is, a tag network map (that is, a hot search tag map) is constructed based on hot search tags, and then a solution is used to improve the quality of search recall results based on the tag network map. The advantage of this embodiment is that after the hot search tags (based on relevance) are mapped, they are applied to the search result recall field, which improves the recall of search results. At the same time, the hot search tags enhance the timeliness of search results.

[0102] In the search, if the keywords searched by the user have too few hits, multiple searches are required to find the right target. How to reduce the number of searches by the user so that the target results can be found quickly becomes the first problem that needs to be solved.

[0103] Based on the first problem to be solved, the search result recall method of this embodiment is as follows: Figure 8 As shown, the following steps are included:

[0104] Step S702, the customer inputs characters;

[0105] Step S704, word segmentation;

[0106] Input characters for word segmentation, and then input vocabulary;

[0107] Step S706, search;

[0108] The search system searches for the first data source according to the user input words (the result after the input characters are segmented);

[0109] Step S708, determining whether the first data source is less than a first threshold;

[0110] If the retrieved first data sources are less than N (N is a first threshold value, which can be defined, such as 10, 20, 30, etc.), the process goes to step S710, otherwise, the retrieval results are directly displayed;

[0111] Step S710, enabling relevant tags and performing retrieval;

[0112] Call the user's input vocabulary to query related hot search tags, enable related tags (hot search tags) as search conditions, perform secondary query, obtain the second data source, and return the results;

[0113] Step S712, the first data source and the second data source are sorted twice;

[0114] The second data source participates in secondary sorting, the sorting weight is customized (can be set according to the label correlation coefficient), and the sorted results are displayed.

[0115] Relevance is an important factor in search results. If you need to add timeliness factors to search results, there are generally two ways: First, by limiting the time of the information source, for example, only returning data before or after a specific date, or weighting the release time of the hit results. The longer the time, the lower the weight. Second, after returning the results, prompt the user to filter the desired data according to the time parameter. However, the above two methods solve the timeliness of the information source itself, but cannot solve the timeliness problem from the search input link.

[0116] Based on the second problem to be solved, the search result recall method of this embodiment is as follows: Fig. 9 As shown, the following steps are included:

[0117] Step S802, collecting hot search tags of the same search object;

[0118] The search object can be a certain product, a certain industry or a usage ID;

[0119] Step S804, obtaining a hot search tag group;

[0120] Each group has one or more hot search tags;

[0121] Step S806: Based on the presence of multiple hot search tags, the hot search tag group is subjected to co-occurrence probability calculation in the form of word vectors to obtain the co-occurrence coefficient of any two hot search tags, i.e., the tag correlation coefficient, and the numerous hot search tags form a tag map;

[0122] In this embodiment, by constructing a hot search tag graph, the correlation coefficient between any two hot search tags is calculated. Because the hot search tags themselves contain the "simultaneous" feature of being simultaneously targeted at a certain search target, the coefficient of any two tags formed by the hot search tag graph contains the time synchronization of the two tags appearing at the same time.

[0123] In this embodiment, a hot search tag map is constructed by adopting an algorithm for the co-occurrence probability of any two words in a word vector.

[0124] This embodiment uses the hot search tag map to assist in improving the search quality. Specifically, first, this embodiment enables related tag (hot search tag) retrieval, which solves the problem of insufficient search results from a semantic level. Second, the hot search tag map (based on co-occurrence probability) solves the user's time requirement at a specific moment.

[0125] In this embodiment, a correlation coefficient map of hot search word labels is constructed based on a co-occurrence probability algorithm, forming a label map containing time synchronization information. During the retrieval process, based on threshold judgment, relevant labels are enabled for secondary retrieval at an appropriate time, which increases simultaneous and relevant recall results and improves the recall quality of the search.

[0126] For example, a diagram of a hot search tag map is shown below. Fig.10 As shown, the vocabulary and correlation coefficients of tags related to "milk" (hot search tags):

[0127] [('Fresh milk',0.8479543924331665),('Yogurt',0.8338502645492554),('Yogurt machine',0.7891680598258972),('Yogurt',0.7624543905258179),('Fresh milk',0.7474728226661682), ('soy milk',0.7275533676147461),('fermentation machine',0.7269923090934753),('goat milk',0.7198804616928101),('peanut milk',0.7125953435897827),('pure milk',0.6916921734809875)].

[0128] The vocabulary and correlation coefficients of the tags related to "peanut milk" (hot search tags):

[0129] [('Almond dew', 0.8878495693206787),('Eight Treasures Porridge', 0.8572490811347961),('Walnut dew', 0.8350837230682373),('Eight Treasures Porridge 320g', 0.8204648494720459),('Eight Treasures Porridge Filling', 0.8162479400634766),( 'Eight Treasures Porridge Production',0.8003721237182617),('Eight Treasures Porridge Ingredients',0.796991765499115),('Eight Treasures Porridge Spoon',0.794583261013031),('Eight Treasures Porridge Agency Franchise',0.7873603105545044),('Eight Treasures Porridge Company',0.7869145274162292)].

[0130] The vocabulary and correlation coefficients of the tags related to "almond milk" (hot search tags):

[0131] [('Peanut milk', 0.8878496289253235),('Peanut milk', 0.886368989944458),('Walnut milk', 0.8728622198104858),('Walnut milk', 0.859451949596405),('Peanut dew', 0.8524070978164673),('Walnut dew ',0.8435949087142944),('Lulu Almond Dew',0.8098096251487732),('Red Date Peanut Dew',0.8036477565765381),('Eight Treasures Porridge 320g',0.8032447099685669),('Eight Treasures Porridge Canned',0.7999293804168701)].

[0132] Embodiment 12:

[0133] The search result recall method of this embodiment is as follows: Fig.11 As shown, the following steps are included:

[0134] Step S902, crawling hot search words for similar products on a certain B2B website to construct hot search phrases.

[0135] Hot search phrases such as Fig.12 shown.

[0136] Step S904, construct word vectors based on the hot search phrases, and each phrase is regarded as a group of word vectors.

[0137] Step S906, calculating the label correlation coefficient based on the word vector.

[0138] Step S908, testing the query results of the hot search tags: querying the correlation coefficient between any two words, and querying the most relevant vocabulary group for any word.

[0139] Query the correlation coefficient of any two words such as Fig.13 As shown, the most relevant vocabulary group for any word is Fig.14 shown.

[0140] Step S910, supplementing the recall results based on the hot search tags.

[0141] In summary, the beneficial effects of the embodiments of the present invention are:

[0142] 1. In this embodiment, by setting a first threshold, when there are too few search results based on the words (keywords) input by the user, the hot search tags corresponding to the words input by the user are obtained, and a secondary search is performed based on the hot search tags. This avoids the problem that when there are too few search results based on the keywords, the user needs to repeatedly try different keywords to search. This solves the problem of insufficient search results from a semantic level, effectively reduces the number of searches by the user, quickly finds the target results, and improves the user experience.

[0143] 2. In this embodiment, when the hot search tag group includes multiple hot search tags, the co-occurrence probability is calculated for the hot search tag group in a word vector manner to obtain the correlation coefficient. By setting the second threshold, it is possible to screen out whether two hot search tags are related, and then obtain a hot search tag map. By adopting the form of a hot search tag map, the timeliness and relevance of the search input are ensured.

[0144] 3. In this embodiment, the hot search tags are mapped and applied to the search result recall field, which improves the quality of the search recall results. At the same time, the hot search tags enhance the timeliness of the search results.

[0145] In the present invention, the terms "first", "second", and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance; the term "plurality" refers to two or more, unless otherwise clearly defined. The terms "installed", "connected", "connected", "fixed", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; "connected" can be a direct connection or an indirect connection through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0146] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by terms such as “upper”, “lower”, “left”, “right”, “front” and “back” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or unit referred to must have a specific direction, be constructed and operated in a specific orientation, and therefore, cannot be understood as a limitation on the present invention.

[0147] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "specific embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0148] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A search result recall method, characterized in that: include: Get the user input characters; Segmenting the user input characters to obtain user input vocabulary; Perform a search based on the user input vocabulary to obtain a first data source; Based on the number of the first data sources being less than a first threshold, obtaining a hot search tag corresponding to the user input word, performing a secondary search based on the hot search tag to obtain a second data source, and displaying the first data source and the second data source; The obtaining of the hot search tag corresponding to the user input word specifically includes: According to the hot search tag map, obtain the hot search tag corresponding to the user input word; Before executing the method of obtaining the user input characters, the method further includes: Collect the hot search tags of the search object, where the hot search tags are one or more; Setting a hot search tag group, wherein the hot search tag group includes all the hot search tags; According to the hot search tag group, construct the hot search tag map; The step of constructing the hot search tag graph according to the hot search tag group specifically includes: Based on the fact that the hot search tag group includes a plurality of the hot search tags, the hot search tag group is subjected to a co-occurrence probability calculation in a word vector manner to obtain a correlation coefficient between any two of the hot search tags, and the two hot search tags having a correlation coefficient greater than a second threshold are set to be correlated to form the hot search tag map; The hot search tag includes a "simultaneous" feature for the user input vocabulary.

2. The search result recall method according to claim 1, characterized in that: The search objects include product names, industry names and / or identity identification numbers.

3. The search result recall method according to claim 1, characterized in that: The step of constructing the hot search tag graph according to the hot search tag group specifically includes: Based on the fact that the hot search tag group includes one of the hot search tags, one of the hot search tags is set as the hot search tag map.

4. The search result recall method according to any one of claims 1 to 3, characterized in that: The displaying the first data source and the second data source specifically includes: The first data source and the second data source are sorted according to the sorting weights and then displayed.

5. The search result recall method according to claim 4, characterized in that: The sorting weight is set according to the correlation coefficient of the hot search tag.

6. A search result recall device (200), characterized in that: include: A memory (210) storing programs or instructions; A processor (220) for executing the program or instruction; Wherein, when executing the program or instruction, the processor (220) implements the steps of the search result recall method as described in any one of claims 1 to 5.

7. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the search result recall method according to any one of claims 1 to 5 are implemented.

Images