Text information pushing method and device thereof

Abstract

The present application proposes a text information pushing method and its device, which relates to the field of Internet technologies. The present application obtains a keyword group composed of at least two keywords and at least one candidate text information; obtains the text information content of the candidate text information in each text information composition dimension; for each candidate text information, obtains the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, obtains the pushing parameter of the candidate text information; and pushes the candidate text information to the terminal device based on the pushing parameter. The present application considers the position and hit count of the keywords in the text information, sets different weight values for each text information composition dimension, so that the calculated pushing parameter can more accurately reflect the matching degree between the candidate text information and the keyword group, and thus can be more accurately pushed to the terminal device.

Description

Technical Field

[0001] The present application relates to the field of Internet technologies, and in particular, to a method and apparatus for pushing text information. Background Art

[0002] When screening text information by searching for keywords, semantic analysis of the text information and keywords is usually used to determine the relevance between the text information and the keywords. However, high accuracy of semantic similarity requires semantic learning of a large number of samples, resulting in a high computational cost. Summary of the Invention

[0003] The present application aims to at least partly solve one of the technical problems in the related art.

[0004] To this end, one object of the present application is to propose a method for pushing text information, which includes obtaining a keyword group composed of at least two keywords and at least one candidate text information; obtaining the text information content of the candidate text information in each text information composition dimension; for each candidate text information, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, obtaining the push parameter of the candidate text information; and pushing the candidate text information to a terminal device based on the push parameter.

[0005] In addition to considering the keyword hit count, the present application also takes into account the position of the keywords in the text information, sets different weight values for each text information composition dimension, and calculates the push parameter of the text information, so that the calculated push parameter can more accurately reflect the matching degree between the candidate text information and the keyword group, and thus the text information pushed to the terminal device is more accurate.

[0006] The second object of the present application is to propose a text information pushing apparatus.

[0007] The third object of the present application is to propose an electronic device.

[0008] The fourth object of the present application is to propose a non-transitory computer-readable storage medium.

[0009] The fifth object of the present application is to propose a computer program product.

[0010] To achieve the above object, an embodiment of the first aspect of the present application provides a method for pushing text information, including: obtaining a keyword group composed of at least two keywords and at least one candidate text information; obtaining the text information content of the candidate text information in each text information composition dimension; for each candidate text information, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, obtaining the push parameter of the candidate text information; and pushing the candidate text information to a terminal device based on the push parameter.

[0011] In addition to considering the keyword hit count, the present application also takes into account the position of the keyword in the text information, sets different weight values for each text information composition dimension, calculates the push parameter of the text information, so that the calculated push parameter can more accurately reflect the matching degree between the candidate text information and the keyword group, and thus the text information pushed to the terminal device is more accurate.

[0012] According to an embodiment of the present application, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group includes: obtaining the second hit count of the text information content corresponding to the text information composition dimension being hit by the keyword group; obtaining the attribution relationship between the text information composition dimensions; and according to the attribution relationship, removing duplicates from the second hit counts corresponding to each text information composition dimension to obtain the first hit count.

[0013] According to an embodiment of the present application, removing duplicates from the second hit counts corresponding to each text information composition dimension according to the attribution relationship to obtain the first hit count includes: based on the attribution relationship, performing hierarchical sorting on all text information composition dimensions to obtain N levels of the text information composition dimensions, where N is a positive integer greater than or equal to 2; obtaining the second hit counts of each text information composition dimension; for any text information composition dimension from the first level to the N-1 level, obtaining the next-level text information composition dimension of the any text information composition dimension, and subtracting the second hit count corresponding to the next-level text information composition dimension from the second hit count corresponding to the any text information composition dimension to obtain the first hit count of the any text information composition dimension.

[0014] According to an embodiment of the present application, the text information composition dimension includes: text information identification text, first-granularity text fragments, second-granularity text fragments, and text information body, where the first-granularity text fragments are used to form the second-granularity text fragments.

[0015] According to an embodiment of the present application, the text information pushing method further includes: in response to any one of the text information composition dimensions being the text information body, subtracting the second hit count of the second-granularity text segment from the second hit count of the text information body to obtain the first hit count of the text information body.

[0016] According to an embodiment of the present application, the text information pushing method further includes: in response to any one of the text information composition dimensions being the second-granularity text segment, subtracting the second hit count of the first-granularity text segment from the second hit count of the second-granularity text segment to obtain the first hit count of the second-granularity text segment.

[0017] According to an embodiment of the present application, the text information pushing method further includes: in response to any one of the text information composition dimensions being the text information identification text, if it is determined according to the attribution relationship that there is no attribution relationship between the text information identification text and the remaining text information composition dimensions, determining the second hit count of the text information identification text as the first hit count of the text information identification text; or, in response to any one of the text information composition dimensions being the first-granularity text segment, if it is determined according to the attribution relationship that there is no next-level text information composition dimension for the first-granularity text segment, determining the second hit count of the first-granularity text segment as the first hit count of the first-granularity text segment.

[0018] According to an embodiment of the present application, the text information pushing method further includes: matching the keywords in the keyword group with the text information to be matched; in response to the text information to be matched containing all the keywords in the keyword group, determining the text information to be matched as the candidate text information.

[0019] According to an embodiment of the present application, obtaining the second hit count of each text information composition dimension includes: for any one of the text information composition dimensions, in response to all the keywords in the keyword group appearing simultaneously in the text information content of the any one of the text information composition dimensions, determining that the text information content of the any one of the text information composition dimensions is hit; counting the number of times the text information content of the any one of the text information composition dimensions is hit by the keyword group to obtain the second hit count of the any one of the text information composition dimensions.

[0020] To achieve the above object, an embodiment of the second aspect of the present application provides a text information pushing device, including: a text information acquisition module, configured to acquire a keyword group composed of at least two keywords and at least one candidate text information; a content acquisition module, configured to acquire the text information content of the candidate text information in each text information composition dimension; a parameter acquisition module, configured to, for each candidate text information, acquire the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, acquire the pushing parameter of the candidate text information; a text information pushing module, configured to push the candidate text information to a terminal device based on the pushing parameter.

[0021] According to an embodiment of the present application, the parameter acquisition module is further configured to: acquire the second hit count of the text information content corresponding to the text information composition dimension being hit by the keyword group; acquire the attribution relationship between the text information composition dimensions; and according to the attribution relationship, remove duplicates from the second hit counts corresponding to each text information composition dimension to obtain the first hit count.

[0022] According to an embodiment of the present application, the parameter acquisition module is further configured to: based on the attribution relationship, perform hierarchical sorting on all the text information composition dimensions to obtain N levels of the text information composition dimensions, where N is a positive integer greater than or equal to 2; acquire the second hit counts of each text information composition dimension; for any text information composition dimension from the first level to the N-1th level, acquire the next-level text information composition dimension of the any text information composition dimension, and subtract the second hit count corresponding to the next-level text information composition dimension from the second hit count corresponding to the any text information composition dimension to obtain the first hit count of the any text information composition dimension.

[0023] According to an embodiment of the present application, the text information composition dimensions include: a text information identification text, a first granularity text segment, a second granularity text segment, and a text information body, where the first granularity text segment is used to form the second granularity text segment.

[0024] According to an embodiment of the present application, the parameter acquisition module is further configured to: in response to the any text information composition dimension being the text information body, subtract the second hit count of the second granularity text segment from the second hit count of the text information body to obtain the first hit count of the text information body.

[0025] According to an embodiment of the present application, the parameter acquisition module is further configured to: in response to any text information composition dimension being the second granularity text segment, subtract the second hit count of the second granularity text segment from the second hit count of the first granularity text segment to obtain the first hit count of the second granularity text segment.

[0026] According to an embodiment of the present application, the parameter acquisition module is further configured to: in response to any text information composition dimension being the text information identification text, if it is determined according to the attribution relationship that there is no attribution relationship between the text information identification text and the remaining text information composition dimensions, determine the second hit count of the text information identification text as the first hit count of the text information identification text; or, in response to any text information composition dimension being the first granularity text segment, if it is determined according to the attribution relationship that there is no next-level text information composition dimension for the first granularity text segment, determine the second hit count of the first granularity text segment as the first hit count of the first granularity text segment.

[0027] According to an embodiment of the present application, the text information acquisition module is further configured to: match the keywords in the keyword group with the text information to be matched; in response to the text information to be matched containing all the keywords in the keyword group, determine the text information to be matched as the candidate text information.

[0028] According to an embodiment of the present application, the parameter acquisition module is further configured to: for any text information composition dimension, if all the keywords in the keyword group appear simultaneously in the text information content of the any text information composition dimension, determine that the text information content of the any text information composition dimension is hit; count the number of times the text information content of the any text information composition dimension is hit by the keyword group to obtain the second hit count of the any text information composition dimension.

[0029] To achieve the above object, an embodiment of the third aspect of the present application provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the text information pushing method as described in the embodiment of the first aspect of the present application is implemented.

[0030] To achieve the above object, an embodiment of the fourth aspect of the present application provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the text information pushing method as described in the embodiment of the first aspect of the present application.

[0031] To achieve the above object, an embodiment of the fifth aspect of the present application provides a computer program product, including a computer program which, when executed by a processor, implements the text information pushing method as described in the embodiment of the first aspect of the present application. Description of the Drawings

[0032] Figure 1 It is a schematic diagram of a text information pushing method according to an embodiment of the present application.

[0033] Figure 2 It is a schematic diagram of obtaining the first hit count of the text information content under each text information composition dimension being hit by a keyword group according to an embodiment of the present application.

[0034] Figure 3 It is a schematic diagram of deduplicating the corresponding second hit counts under each text information composition dimension according to an embodiment of the present application.

[0035] Figure 4 It is a general flowchart of a text information pushing method according to an embodiment of the present application.

[0036] Figure 5 It is a schematic diagram of a text information pushing device according to an embodiment of the present application.

[0037] Figure 6 It is a schematic diagram of an electronic device according to an embodiment of the present application. Detailed Embodiments

[0038] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present application, but should not be construed as limiting the present application.

[0039] Figure 1 It is an exemplary implementation of a text information pushing method of the present application. As Figure 1 shown, the text information pushing method includes the following steps:

[0040] S101, obtain a keyword group composed of at least two keywords and at least one candidate text information.

[0041] When obtaining candidate text information, the candidate text information can be pre-determined text information, or can be obtained by matching keyword groups with the text information to be matched. For example, the candidate text information can be news, articles, etc. that need to be matched with keyword groups. Among them, the keyword group contains at least two keywords. The keywords can be words entered by the user in the search box, or keywords extracted according to the keyword extraction algorithm from the first-granularity text fragment entered by the user in the search box. Exemplarily, the first-granularity text fragment can be a sentence, that is, the user can directly enter a sentence in the search box to obtain the keywords in the sentence. Optionally, the keyword extraction algorithm can adopt the Natural Language Processing (NLP) algorithm.

[0042] When setting the keyword group, generally, the scope defined by the first keyword in the keyword group is larger, and the second and subsequent keywords can focus on specific things, brands, products, etc., so that the matched candidate text information is more focused and has a stronger relevance.

[0043] S102, obtain the text information content of the candidate text information in each dimension of the text information composition.

[0044] Considering that the occurrence positions of keywords in candidate text information are often different, they may appear in the first-granularity text fragment, may appear in different first-granularity text fragments in the second-granularity text fragment, or may appear in different second-granularity text fragments in the text information body. For ease of understanding, it can be exemplified that the first-granularity text fragment can be a sentence, and the second-granularity text fragment can be a paragraph, that is, a paragraph can include multiple sentences. The keywords appear in different positions, and the influence on the relevance of the candidate text information is different. Taking the keyword group containing two keywords as an example, if both of these two keywords appear in the text marked by a certain candidate text information, then this candidate text information is probably written around these two keywords, that is, this candidate text information is closely related to these two keywords; if one of the keywords in the keyword group appears in the first second-granularity text fragment of the text information, and the other keyword in the keyword group appears in the last second-granularity text fragment, and none of the keywords in the keyword group appear in the remaining part, then the probability that this candidate text information is closely related to these two keywords is smaller.

[0045] Considering that the impacts caused by the different positions of keywords in the candidate text information are different, the candidate text information can be split into different dimensions of text information components. Optionally, the dimensions of text information components may include: text information identification text, first-granularity text fragments, second-granularity text fragments, and text information body, where the first-granularity text fragments are used to form the second-granularity text fragments. Among them, the second-granularity text fragments are obtained by splitting the text information body, and the first-granularity text fragments are obtained by splitting the second-granularity text fragments. The text information identification text is not split. It should be noted that the text information body does not include the text information identification text. It should be noted that splitting the candidate text information into text information identification text, first-granularity text fragments, second-granularity text fragments, and text information body here is only an example and cannot be used as a condition for restricting this application.

[0046] In this application, in order to facilitate the statistics of the hit times of keyword groups in each dimension of text information components, it is necessary to obtain the text content corresponding to the text information identification text, first-granularity text fragments, second-granularity text fragments, and text information body in the candidate text information. When obtaining the content of the text information identification text, it can be extracted at the target position corresponding to the candidate text information; when obtaining the text information content corresponding to the text information body, the content of the text information identification text that has been obtained can be removed from the entire candidate text information, and then the text information content corresponding to the text information body can be obtained; when obtaining the text information content corresponding to the second-granularity text fragments, based on the format of the candidate text information, for example, when the first-granularity text fragments can be sentences and the second-granularity text fragments can be paragraphs, each second-granularity text fragment will be indented two characters at the beginning of the line to divide the text information body and obtain the text information content corresponding to each second-granularity text fragment; when obtaining the text information content corresponding to the first-granularity text fragments, the text information content of each second-granularity text fragment can be divided based on punctuation marks or semantic integrity.

[0047] S103. For each candidate text information, obtain the first hit times when the text information content in each dimension of text information components is hit by the keyword group, and based on the weight value corresponding to each dimension of text information components and the first hit times corresponding to the dimension of text information components, obtain the push parameter of the candidate text information.

[0048] Obtain the first hit count of the text information content in each dimension of the text information corresponding to each candidate text information being hit by the keyword group. It should be noted that when counting the hit count of the keyword group in the text information content in each dimension of the text information composition, there will be duplicate counting situations. For example, if a keyword group appears in a first-granularity text segment and the hit count of the keyword group is counted, then the same keyword group also appears in the second-granularity text segment corresponding to this first-granularity text segment and is counted, and the same keyword group also appears in the text information body corresponding to this second-granularity text segment and is counted, resulting in the same keyword group being counted more than once. To avoid duplicate counting of the keyword group, when counting the hit count of the keyword group, it is necessary to deduplicate the hit count based on the position where the keyword group appears, so as to obtain the deduplicated first hit count.

[0049] Considering that the impacts caused by the different positions of the keyword in the candidate text information are different, optionally, different weight values can be set for each dimension of the text information composition corresponding to the candidate text information according to the influence level. Based on the weight value corresponding to each dimension of the text information composition set in advance and the first hit count corresponding to the dimension of the text information composition, multiply the weight value corresponding to each dimension of the text information composition by the first hit count corresponding to this dimension of the text information composition respectively, and take the sum of each product as the push parameter of the candidate text information.

[0050] Table 1 is an exemplary relationship table of the weight value and the first hit count corresponding to the dimension of the text information composition of a certain candidate text information. As shown in Table 1, taking the first hit counts of the text information identification text, the first-granularity text segment, the second-granularity text segment, and the text information body as 1, 5, 3, and 4 respectively, the weight values of the text information identification text, the first-granularity text segment, the second-granularity text segment, and the text information body are set to 0.4, 0.3, 0.2, and 0.1 respectively. The push parameter of this candidate text information is: 1×0.4 + 5×0.3 + 3×0.2 + 4×0.1 = 2.9.

[0051] Table 1

[0052] Composition Dimensions of Text Information First Hit Count Weight Value Text Identifying Text of Text Information 1 0.4 First-Granularity Text Fragment 5 0.3 Second-Granularity Text Fragment 3 0.2 Main Body of Text Information 4 0.1

[0053] S104, based on the push parameter, push the candidate text information to the terminal device.

[0054] Sort the obtained push parameters in descending order, and thus sort the candidate text information corresponding to each push parameter according to the order of the push parameters.

[0055] As an implementable manner, when pushing candidate text information to a terminal device, all candidate text information may be sequentially pushed to the terminal device in the order of the candidate text information. Optionally, the terminal device may be a mobile phone, a tablet, etc.

[0056] As another implementable manner, when pushing candidate text information to a terminal device, a preset number of candidate text information may be sequentially pushed to the terminal device in the order of the candidate text information. For example, if the preset number is 50, when pushing candidate text information to the terminal device, the first 50 candidate text information may be sequentially pushed to the terminal device in order. Optionally, the terminal device may be a mobile phone, a tablet, etc.

[0057] An embodiment of the present application proposes a method for pushing text information, which includes obtaining a keyword group composed of at least two keywords and at least one candidate text information; obtaining the text information content of the candidate text information in each text information composition dimension; for each candidate text information, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, obtaining the push parameter of the candidate text information; and pushing the candidate text information to the terminal device based on the push parameter. In addition to the keyword hit count, the present application also takes into account the position of the keyword in the text information, sets different weight values for each text information composition dimension, calculates the push parameter of the text information, so that the calculated push parameter can more accurately reflect the matching degree between the candidate text information and the keyword group, and thus the text information pushed to the terminal device is more accurate.

[0058] Figure 2 is an exemplary implementation manner of a method for pushing text information in the present application. As Figure 2 shown, based on the above embodiment, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group includes the following steps:

[0059] S201, obtaining the second hit count of the text information content corresponding to the text information composition dimension being hit by the keyword group.

[0060] According to the above, considering that the positions of the keywords in the candidate text information are different, the candidate text information may be split into different text information composition dimensions. Optionally, the text information composition dimension may include: text information identification text, first granularity text segment, second granularity text segment, and text information body, where the first granularity text segment is used to form the second granularity text segment.

[0061] Match the text information content corresponding to each dimension of the text information with all the keywords in the keyword group respectively. Among them, for any dimension of the text information composition, when all the keywords in the keyword group appear simultaneously in the text information content of any dimension of the text information composition, it is determined that the text information content of this dimension of the text information composition is hit, and count the number of times the text information content of each dimension of the text information composition is hit by the keyword group to obtain the second hit count of each dimension of the text information composition.

[0062] For example, if the keyword group is "fruit" + "watermelon", in the text information content of any dimension of the text information composition, if both the keywords "fruit" and "watermelon" appear simultaneously, it is determined that the text information content of this dimension of the text information composition is hit, and the second hit count corresponding to this dimension of the text information composition is incremented by 1. If only the keyword "fruit" appears in the text information content of any dimension of the text information composition and the keyword "watermelon" does not appear, it is determined that this dimension of the text information composition is not hit by the keyword group, and the second hit count corresponding to this dimension of the text information composition is not incremented.

[0063] It should be noted that when counting the second hit count for any dimension of the text information composition, since the number of times each keyword in the keyword group appears may be different, in this application, the number of times each keyword appears can be sorted, and the minimum value of the number of times each keyword appears is selected as the second hit count corresponding to this dimension of the text information composition.

[0064] For example, if the keyword group is "fruit" + "watermelon", in the dimension of the second-granularity text segment of the text information composition, both the keywords "fruit" and "watermelon" appear simultaneously, but the keyword "fruit" appears 5 times and the keyword "watermelon" appears 3 times, then the second hit count corresponding to the second-granularity text segment is 3 times.

[0065] The statistical method for the second hit count of the first-granularity text segment, the text information identification text, and the text information body is the same as the statistical method for the second hit count of the above-mentioned second-granularity text segment, and will not be elaborated here.

[0066] S202, obtain the attribution relationship between dimensions of the text information.

[0067] As described above, the second-granularity text fragments are obtained by splitting the body of the text information, and the first-granularity text fragments are obtained by splitting the second-granularity text fragments. The text information identification text is not split. It can be seen that there is an inclusion relationship among the composition dimensions of the text information. Among them, the body of the text information contains the second-granularity text fragments, the second-granularity text fragments contain the first-granularity text fragments, and there is no inclusion relationship between the text information identification text and each of the remaining composition dimensions of the text information. That is to say, the text information identification text does not belong to any of the remaining composition dimensions of the text information.

[0068] S203. According to the inclusion relationship, deduplicate the second hit counts corresponding to each composition dimension of the text information to obtain the first hit counts.

[0069] As Figure 3 shown, when deduplicating the second hit counts corresponding to each composition dimension of the text information according to the inclusion relationship to obtain the first hit counts, the following steps are included:

[0070] S301. Based on the inclusion relationship, perform a hierarchical sorting on all the composition dimensions of the text information to obtain N levels of the composition dimensions of the text information, where N is a positive integer greater than or equal to 2.

[0071] According to the inclusion relationship among the composition dimensions of the text information obtained above, perform a hierarchical sorting on multiple composition dimensions of the text information, which are respectively denoted as the 1st level to the Nth level. Among them, in the hierarchical order from low to high, the composition dimensions of the text information can be sorted as: the first-granularity text fragments, the second-granularity text fragments, and the body of the text information. In addition, the composition dimensions of the text information also include the text information identification text, and there is no inclusion relationship between the text information identification text and each of the remaining composition dimensions of the text information.

[0072] S302. Obtain the second hit counts of each composition dimension of the text information.

[0073] Regarding obtaining the second hit counts of each composition dimension of the text information, specific introduction has been made in S201 above, and will not be elaborated here.

[0074] S303. For any composition dimension of the text information from the 1st level to the (N - 1)th level, obtain the next-level composition dimension of any composition dimension of the text information, and subtract the second hit count corresponding to the next-level composition dimension of any composition dimension of the text information from the second hit count corresponding to any composition dimension of the text information to obtain the first hit count of any composition dimension of the text information.

[0075] Due to the attribution relationship among the constituent dimensions of text information, there will be a situation where the keyword groups are counted repeatedly in the text information content of the lower-level text information constituent dimension and the text information content of the higher-level text information constituent dimension. Therefore, it is necessary to remove duplicates from the corresponding second hit counts under each text information constituent dimension.

[0076] As an implementable way, when the text information constituent dimension is the text information body, the first hit count of the text information body is obtained by subtracting the second hit count of the second granularity text segment from the second hit count of the text information body.

[0077] As another implementable way, when the text information constituent dimension is the second granularity text segment, the first hit count of the second granularity text segment is obtained by subtracting the second hit count of the first granularity text segment from the second hit count of the second granularity text segment.

[0078] As another implementable way, when the text information constituent dimension is the text information identification text, according to the attribution relationship, it is determined that there is no attribution relationship between the text information identification text and the remaining text information constituent dimensions, and the second hit count of the text information identification text is determined as the first hit count of the text information identification text.

[0079] As another implementable way, when the text information constituent dimension is the first granularity text segment, according to the attribution relationship, it is determined that there is no lower-level text information constituent dimension for the first granularity text segment, and the second hit count of the first granularity text segment is determined as the first hit count of the first granularity text segment.

[0080] Table 2 is an exemplary relationship table of the first hit count and the second hit count of the text information constituent dimension. As shown in Table 2, the first hit count of the text information body is equal to the second hit count of the text information body minus the second hit count of the second granularity text segment; the first hit count of the second granularity text segment is equal to the second hit count of the second granularity text segment minus the second hit count of the first granularity text segment; the first hit count and the second hit count of the first granularity text segment are the same; the first hit count and the second hit count of the text information identification text are the same.

[0081] Table 2

[0082] Composition Dimensions of Text Information Second Hit Count First Hit Count Text Identifying Text of Text Information 1 1 First-Granularity Text Fragment 5 5 Second-Granularity Text Fragment 8 3 Main Body of Text Information 12 4

[0083] In the embodiments of the present application, according to the attribution relationship, duplicates are removed from the corresponding second hit counts under each text information constituent dimension to obtain the first hit count, which facilitates the subsequent calculation of the push parameters and makes the calculated push parameters more accurately reflect the matching degree between the candidate text information and the keywords.

[0084] Figure 4 This is an exemplary embodiment of a method for pushing text information in the present application. As Figure 4 shown, the method for pushing text information includes the following steps:

[0085] S401, Obtain a keyword group composed of at least two keywords and at least one candidate text information.

[0086] S402, Obtain the text information content of the candidate text information in each dimension of the text information composition.

[0087] Regarding S401 to S402, the above embodiments have been specifically introduced and will not be elaborated here.

[0088] S403, Obtain the second hit count of the text information content corresponding to the dimension of the text information composition being hit by the keyword group.

[0089] S404, Obtain the attribution relationship between the dimensions of the text information composition.

[0090] S405, Based on the attribution relationship, perform a hierarchical sorting on all dimensions of the text information composition to obtain N levels of the dimensions of the text information composition, where N is a positive integer greater than or equal to 2.

[0091] S406, For any dimension of the text information composition from the first level to the N - 1 level, obtain the next-level dimension of the text information composition of any dimension, and subtract the second hit count corresponding to the next-level dimension of the text information composition from the second hit count corresponding to any dimension of the text information composition to obtain the first hit count of any dimension of the text information composition.

[0092] Regarding S403 to S406, the above embodiments have been specifically introduced and will not be elaborated here.

[0093] S407, Based on the weight value corresponding to each dimension of the text information composition and the first hit count corresponding to the dimension of the text information composition, obtain the push parameter of the candidate text information.

[0094] S408, Based on the push parameter, push the candidate text information to the terminal device.

[0095] An embodiment of the present application provides a method for pushing text information. The method includes: obtaining a keyword group composed of at least two keywords and at least one candidate text information; obtaining the text information content of the candidate text information in each text information composition dimension; for each candidate text information, obtaining a first hit count of the text information content in each text information composition dimension being hit by the keyword group, and obtaining a push parameter of the candidate text information based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension; and pushing the candidate text information to a terminal device based on the push parameter. In addition to the keyword hit count, the present application also takes into account the position of the keyword in the text information, sets different weight values for each text information composition dimension, calculates the push parameter of the text information, so that the calculated push parameter can more accurately reflect the matching degree between the candidate text information and the keyword group, and thus the text information pushed to the terminal device is more accurate.

[0096] Figure 5 is an exemplary schematic diagram of a text information pushing device proposed by the present application, as Figure 5 shown. The text information pushing device 500 includes: a text information obtaining module 51, a content obtaining module 52, a parameter obtaining module 53, and a text information pushing module 54, where:

[0097] The text information obtaining module 51 is configured to obtain a keyword group composed of at least two keywords and at least one candidate text information.

[0098] The content obtaining module 52 is configured to obtain the text information content of the candidate text information in each text information composition dimension.

[0099] The parameter obtaining module 53 is configured to, for each candidate text information, obtain a first hit count of the text information content in each text information composition dimension being hit by the keyword group, and obtain a push parameter of the candidate text information based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension.

[0100] The text information pushing module 54 is configured to push the candidate text information to a terminal device based on the push parameter.

[0101] Further, the parameter obtaining module 53 is further configured to: obtain a second hit count of the text information content corresponding to the text information composition dimension being hit by the keyword group; obtain the attribution relationship between the text information composition dimensions; and de-duplicate the second hit counts corresponding to each text information composition dimension according to the attribution relationship to obtain the first hit count.

[0102] Further, the parameter acquisition module 53 is further configured to: hierarchically sort all the text information composition dimensions based on the attribution relationship to obtain N levels of the text information composition dimensions, where N is a positive integer greater than or equal to 2; obtain the second hit counts of each text information composition dimension; for any text information composition dimension from the first level to the N-1th level, obtain the next-level text information composition dimension of any text information composition dimension, and subtract the second hit count corresponding to the next-level text information composition dimension from the second hit count corresponding to any text information composition dimension to obtain the first hit count of any text information composition dimension.

[0103] Further, the text information composition dimensions include: text information identification text, first-granularity text fragments, second-granularity text fragments, and text information body, where the first-granularity text fragments are used to form the second-granularity text fragments.

[0104] Further, the parameter acquisition module 53 is further configured to: in response to any text information composition dimension being the text information body, subtract the second hit count of the text information body from the second hit count of the second-granularity text fragments to obtain the first hit count of the text information body.

[0105] Further, the parameter acquisition module 53 is further configured to: in response to any text information composition dimension being the second-granularity text fragments, subtract the second hit count of the first-granularity text fragments from the second hit count of the second-granularity text fragments to obtain the first hit count of the second-granularity text fragments.

[0106] Further, the parameter acquisition module 53 is further configured to: in response to any text information composition dimension being the text information identification text, determine that there is no attribution relationship between the text information identification text and the remaining text information composition dimensions according to the attribution relationship, and determine the second hit count of the text information identification text as the first hit count of the text information identification text; or, in response to any text information composition dimension being the first-granularity text fragments, determine that there is no next-level text information composition dimension for the first-granularity text fragments according to the attribution relationship, and determine the second hit count of the first-granularity text fragments as the first hit count of the first-granularity text fragments.

[0107] Further, the text information acquisition module 51 is further configured to: match the keywords in the keyword group with the text information to be matched; in response to the text information to be matched containing all the keywords in the keyword group, determine the text information to be matched as the candidate text information.

[0108] Further, the parameter acquisition module 53 is further configured to: for any text information composition dimension, in response to all keywords in the keyword group appearing simultaneously in the text information content of any text information composition dimension, determine that the text information content of any text information composition dimension is hit; count the number of times the text information content of any text information composition dimension is hit by the keyword group, and obtain the second hit count of any text information composition dimension.

[0109] To implement the above embodiments, an electronic device 600 is further proposed in an embodiment of the present application. As Figure 6 shown, the electronic device 600 includes: a processor 601 and a memory 602 communicatively connected to the processor. The memory 602 stores instructions executable by at least one processor. The instructions are executed by at least one processor 601 to implement the text information pushing method as shown in the above embodiments.

[0110] To implement the above embodiments, a non-transitory computer-readable storage medium storing computer instructions is further proposed in an embodiment of the present application, wherein the computer instructions are used to cause a computer to implement the text information pushing method as shown in the above embodiments.

[0111] To implement the above embodiments, a computer program product is further proposed in an embodiment of the present application, including a computer program, where the computer program implements the text information pushing method as shown in the above embodiments when executed by a processor.

[0112] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present application.

[0113] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, "a plurality" means two or more, unless otherwise specifically defined.

[0114] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples.

[0115] Although the embodiments of this application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting this application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for pushing text information, characterized in that, Including: Obtaining a keyword group composed of at least two keywords and at least one candidate text information; Obtaining the text information content of the candidate text information in each text information composition dimension; For each candidate text information, obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, obtaining the push parameter of the candidate text information; Based on the push parameter, pushing the candidate text information to the terminal device; The obtaining the first hit count of the text information content in each text information composition dimension being hit by the keyword group includes: Obtaining the second hit count of the corresponding text information content in the text information composition dimension being hit by the keyword group, wherein, according to the position of the keyword in the candidate text information, the candidate text information is split into different text information composition dimensions; Obtaining the attribution relationship between the text information composition dimensions; According to the attribution relationship, removing duplicates from the second hit counts corresponding to each text information composition dimension to obtain the first hit count.

2. The method according to claim 1, characterized in that, The removing duplicates from the second hit counts corresponding to each text information composition dimension according to the attribution relationship to obtain the first hit count includes: Based on the attribution relationship, performing hierarchical sorting on all the text information composition dimensions to obtain N levels of the text information composition dimensions, wherein N is a positive integer greater than or equal to 2; Obtaining the second hit count of each text information composition dimension; For any text information composition dimension from the first level to the N-1 level, obtaining the next-level text information composition dimension of the any text information composition dimension, and subtracting the second hit count corresponding to the next-level text information composition dimension from the second hit count corresponding to the any text information composition dimension to obtain the first hit count of the any text information composition dimension.

3. The method according to claim 2, characterized in that, The text information composition dimension includes: text information identification text, first-granularity text fragment, second-granularity text fragment, and text information body, wherein the first-granularity text fragment is used to form the second-granularity text fragment.

4. The method according to claim 3, characterized in that, The method further includes: In response to the any text information composition dimension being the text information body, subtracting the second hit count of the text information body from the second hit count of the second-granularity text fragment to obtain the first hit count of the text information body.

5. The method according to claim 4, characterized in that, The method further includes: In response to the any text information composition dimension being the second-granularity text fragment, subtracting the second hit count of the first-granularity text fragment from the second hit count of the second-granularity text fragment to obtain the first hit count of the second-granularity text fragment.

6. The method according to claim 3, characterized in that, The method further includes: In response to any of the text information composition dimensions being the text identifying the text information, if it is determined according to the attribution relationship that there is no attribution relationship between the text identifying the text information and the remaining text information composition dimensions, the second hit count of the text identifying the text information is determined as the first hit count of the text identifying the text information; or, In response to any of the text information composition dimensions being the first granularity text segment, if it is determined according to the attribution relationship that there is no next-level text information composition dimension for the first granularity text segment, the second hit count of the first granularity text segment is determined as the first hit count of the first granularity text segment.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Matching the keywords in the keyword group with the text information to be matched; In response to the text information to be matched containing all the keywords in the keyword group, determining the text information to be matched as the candidate text information.

8. The method according to claim 2, characterized in that, The obtaining the second hit count of each of the text information composition dimensions includes: For any text information composition dimension, if all the keywords in the keyword group appear simultaneously in the text information content of the any text information composition dimension, it is determined that the text information content of the any text information composition dimension is hit; Counting the number of times the text information content of the any text information composition dimension is hit by the keyword group to obtain the second hit count of the any text information composition dimension.

9. A device for pushing text information, characterized in that, It includes: A text information acquisition module, configured to acquire a keyword group composed of at least two keywords and at least one candidate text information; A content acquisition module, configured to acquire the text information content of the candidate text information under each text information composition dimension; A parameter acquisition module, configured to, for each candidate text information, acquire the first hit count of the text information content under each text information composition dimension being hit by the keyword group, and based on the weight value corresponding to each text information composition dimension and the first hit count corresponding to the text information composition dimension, acquire the push parameter of the candidate text information; A text information push module, configured to push the candidate text information to the terminal device based on the push parameter; The parameter acquisition module is further configured to: Obtain the second hit count of the text information content corresponding to the text information composition dimension being hit by the keyword group, where the candidate text information is split into different text information composition dimensions according to the position of the keyword in the candidate text information; Obtain the attribution relationship between the text information composition dimensions; According to the attribution relationship, deduplicate the second hit counts corresponding to each text information composition dimension to obtain the first hit count.

10. The device according to claim 9, wherein, The parameter acquisition module is further configured to: Based on the attribution relationship, perform hierarchical sorting on all the text information composition dimensions to obtain N levels of the text information composition dimensions, where N is a positive integer greater than or equal to 2; Obtain the second hit count of each of the text information composition dimensions; For any text information composition dimension from the first level to the (N - 1)th level, obtain the next-level text information composition dimension of the any text information composition dimension, and subtract the second hit count corresponding to the next-level text information composition dimension from the second hit count corresponding to the any text information composition dimension to obtain the first hit count of the any text information composition dimension.

11. The device according to claim 10, wherein, The text information composition dimension includes: a text information identification text, a first-granularity text segment, a second-granularity text segment, and a text information body, wherein the first-granularity text segment is used to form the second-granularity text segment.

12. The device according to claim 11, wherein, The parameter acquisition module is further configured to: In response to the any text information composition dimension being the text information body, subtract the second hit count of the text information body from the second hit count of the second-granularity text segment to obtain the first hit count of the text information body.

13. The device according to claim 12, wherein, The parameter acquisition module is further configured to: In response to the any text information composition dimension being the second-granularity text segment, subtract the second hit count of the first-granularity text segment from the second hit count of the second-granularity text segment to obtain the first hit count of the second-granularity text segment.

14. The device according to claim 11, wherein, The parameter acquisition module is further configured to: In response to the any text information composition dimension being the text information identification text, determine that there is no attribution relationship between the text information identification text and the remaining text information composition dimensions according to the attribution relationship, and determine the second hit count of the text information identification text as the first hit count of the text information identification text; Or, In response to the any text information composition dimension being the first-granularity text segment, determine that there is no next-level text information composition dimension for the first-granularity text segment according to the attribution relationship, and determine the second hit count of the first-granularity text segment as the first hit count of the first-granularity text segment.

15. The device according to any one of claims 9-14, wherein, The text information acquisition module is further configured to: Match the keywords in the keyword group with the text information to be matched; In response to the text information to be matched including all the keywords in the keyword group, determine the text information to be matched as the candidate text information.

16. The device according to claim 10, wherein, The parameter acquisition module is further configured to: For any text information composition dimension, in response to all the keywords in the keyword group simultaneously appearing in the text information content of the any text information composition dimension, determine that the text information content of the any text information composition dimension is hit; Count the number of times the text information content of the any text information composition dimension is hit by the keyword group to obtain the second hit count of the any text information composition dimension.

17. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1 - 8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are for causing the computer to execute the method according to any one of claims 1-8.

19. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 - 8.

Images