CN108256044B - Recommended method, device and electronic equipment for live broadcast room - Google Patents
Recommended method, device and electronic equipment for live broadcast room Download PDFInfo
- Publication number
- CN108256044B CN108256044B CN201810029064.2A CN201810029064A CN108256044B CN 108256044 B CN108256044 B CN 108256044B CN 201810029064 A CN201810029064 A CN 201810029064A CN 108256044 B CN108256044 B CN 108256044B
- Authority
- CN
- China
- Prior art keywords
- index
- word
- words
- search
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本发明公开了一种直播间推荐方法、装置及电子设备，属于互联网技术领域。所述方法包括：获取用户输入的搜索词的特征词；获取所述特征词与预设的索引词词典中的索引词之间的匹配度，其中，所述索引词词典包括一种或多种索引类别，每种所述索引类别对应有预设匹配规则以及多个所述索引词，每个所述索引词对应有直播间；根据所述匹配度以及每种所述索引类别对应的所述预设匹配规则确定所述搜索词对应的所述索引词；根据所述搜索词对应的所述索引词向所述用户推荐相应的所述直播间。有利于提高直播间推荐结果的可靠性，以提升用户体验。
The invention discloses a method, device and electronic equipment for recommending a live broadcast room, belonging to the technical field of Internet. The method includes: acquiring a characteristic word of a search word input by a user; acquiring a degree of matching between the characteristic word and an index word in a preset index word dictionary, wherein the index word dictionary includes one or more Index categories, each of the index categories corresponds to a preset matching rule and a plurality of the index words, and each of the index words corresponds to a live room; according to the matching degree and the corresponding index categories The index word corresponding to the search word is determined by a preset matching rule; the corresponding live room is recommended to the user according to the index word corresponding to the search word. It is beneficial to improve the reliability of the recommendation results in the live broadcast room, so as to improve the user experience.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a live broadcast room recommendation method and device and electronic equipment.
Background
On a live platform, searching is an important recommendation scene. When a user searches, the user will want to see some specific content. Therefore, it is possible to guess the real intention of the user from the search word of the user, thereby recommending the content related to the real intention thereof.
In the prior art, search terms input by a user are generally directly matched with names of live broadcast rooms, and the live broadcast rooms with the first two characters of the search terms in the names are recommended to the user. Since the live room name is named for the anchor according to its own thoughts. Therefore, only the search terms are matched with the names of the live broadcast rooms, and the obtained recommendation results are low in reliability and poor in relevance with the real intentions of the users.
Disclosure of Invention
In view of the above, the present invention is proposed to provide a live broadcast recommendation method, apparatus and electronic device that overcome or at least partially solve the above problems.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, an embodiment of the present invention provides a live broadcast room recommendation method, where the method includes: acquiring characteristic words of search words input by a user; acquiring the matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast; determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category; and recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
Preferably, the index dictionary includes a first index category, and the step of determining the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category includes: judging whether the matching degree between the feature words and first target index words meets a first preset condition or not, wherein the first target index words comprise index words corresponding to the first index category; and if the first preset condition is met, taking the first target index word with the matching degree meeting the first preset condition as the index word corresponding to the search word.
Preferably, the step of determining the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category includes: judging whether the matching degree between the feature words and second target index words meets a second preset condition or not, wherein the second target index words comprise index words corresponding to the second index category; and if the second preset condition is met, taking the second target index word with the matching degree meeting the second preset condition as the index word corresponding to the search word.
Preferably, the step of judging whether the matching degree between the feature word and the second target index word meets a second preset condition includes: obtaining a characteristic value according to the text length of the characteristic word, the text length of the second target index word, the matching degree between the characteristic word and the second target index word and a preset algorithm; if the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets a second preset condition; and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
Preferably, before obtaining the feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word, and a preset algorithm, the method further includes: acquiring the text length of the feature words; judging whether the text length of the feature words is larger than a preset length threshold value or not; and if the text length of the feature word is larger than the length threshold, executing the step of obtaining a feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word and a preset algorithm.
Preferably, the step of obtaining the matching degree between the feature word and an index word in a preset index word dictionary includes: and acquiring an editing distance between the feature words and index words in a preset index word dictionary, and taking the editing distance as the matching degree between the feature words and the index words.
Preferably, the step of acquiring the feature words of the search words input by the user includes: acquiring a search word input by a user; and performing word segmentation processing on the search word to obtain one or more search words, and taking the one or more search words as the characteristic word.
In a second aspect, an embodiment of the present invention further provides a live broadcast room recommendation apparatus, where the apparatus includes: the device comprises a characteristic word acquisition module, a matching degree acquisition module, an index word determination module and a recommendation module. The characteristic word acquisition module is used for acquiring the characteristic words of the search words input by the user. The matching degree obtaining module is used for obtaining the matching degree between the characteristic words and the index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast room. And the index word determining module is used for determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category. And the recommending module is used for recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a processor and a memory, where the memory is coupled to the processor, and the memory stores instructions. The instructions, when executed by the processor, cause the electronic device to: acquiring characteristic words of search words input by a user; acquiring the matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast; determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category; and recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The program realizes the steps of the live broadcast room recommendation method when being executed by a processor.
In the technical scheme of the embodiment of the invention, the characteristic words of the search words input by a user are firstly obtained, then the matching degree between the characteristic words and the index words in a preset index word dictionary is obtained, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, each index word corresponds to a live broadcast room, then the index words corresponding to the search words are determined according to the matching degree and the preset matching rule corresponding to each index category, and then the corresponding live broadcast rooms are recommended to the user according to the index words corresponding to the search words. Different index categories correspond to different search intentions, and the index words corresponding to the search words are determined by combining the search intentions and the matching degrees, so that the reliability of the recommendation result of the live broadcast room is improved, namely, the relevance between the recommended live broadcast room and the real intention of the user is improved, and the user experience is improved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 shows a flowchart of a live broadcast recommendation method according to a first embodiment of the present invention;
FIG. 2 shows a flow chart of a first embodiment of step S103 of FIG. 1;
FIG. 3 shows a flowchart of a second embodiment of step S103 of FIG. 1;
FIG. 4 shows a flowchart of a third embodiment of step S103 of FIG. 1;
fig. 5 shows a block diagram of a live broadcast recommendation apparatus according to a second embodiment of the present invention;
fig. 6 shows a block diagram of an index word determination module in a live broadcast recommendation apparatus according to a second embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a live broadcast room recommendation method and device and electronic equipment, which can effectively improve the reliability of a live broadcast room recommendation result, namely improve the correlation between a recommended live broadcast room and the real intention of a user so as to improve the user experience. The method comprises the following steps: acquiring characteristic words of search words input by a user; acquiring the matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast; determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category; and recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Referring to fig. 1, a first embodiment of the present invention provides a live broadcast recommendation method. As shown in fig. 1, the method comprises the steps of:
step S101, acquiring characteristic words of search words input by a user;
based on the desire to see some specific content, the user inputs corresponding search terms on the live platform to obtain related live room search results. That is, the search terms may reflect the user's search intent.
As one embodiment, the step of obtaining the feature words of the search words input by the user includes: acquiring a search word input by a user; and performing word segmentation processing on the search word to obtain one or more search words, and taking the one or more search words as the characteristic word.
Of course, before performing the word segmentation processing on the search word, the method may further include: and preprocessing the search word. In this embodiment, the search term may be preprocessed according to specific needs. For example, the pre-processing may include, but is not limited to, one or more of deleting special characters such as "#" and "&", converting pinyin to simplified words, converting traditional words to simplified words, converting full-angle characters to half-angle characters, converting uppercase letters to lowercase letters, and the like.
In this embodiment, one or more feature words of the search term acquired in step S101 may be used. When the feature word is plural, the following steps S102 and S103 need to be performed for each feature value acquired in step S101. For example, a feature word sequence may be constructed by performing word segmentation on a search word, a first feature value in the feature word sequence is used as a current feature value, the following steps S102 and S103 are performed on the current feature value, a next feature value in the feature word sequence is used as the current feature value, and the following steps S102 and S103 are performed until all feature values in the feature word sequence are performed, and then step S104 is performed.
Step S102, obtaining a matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast room;
it is understood that, before step S102 is executed, an index word dictionary needs to be created in advance, such that the index word dictionary includes one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast.
The index categories correspond to the search intention of the user, the index dictionary comprises one or more index categories, and the index categories can be specifically divided according to actual application. For example, in one particular application scenario, a user's search intent on a live platform includes an anchor intent, a partition intent, and a tag intent. Anchor intent refers to a user wishing to search for a particular anchor, partition intent refers to a user wishing to search for related content in a certain partition, and tag intent refers to a user wishing to search for related content under a certain tag. Only if the search intention of the user is accurately identified, the recommendation result can be reasonably returned according to the intention of the user.
The index words corresponding to the index categories in the index word dictionary are corresponding words expressed by the intentions. For example, the corresponding index word is typically a nickname for the anchor under the anchor intent; the index word of the partition intention is a secondary partition of a live platform, such as game A, game B and the like, and synonyms of the partitions; the index words of the label intention are the live room content labels generated according to other methods, such as humorous fun, technical height, and the like, and synonyms of the labels. Therefore, according to the different classification intents, the index words can be divided into three categories, namely, anchor index words, partition index words and tag index words. At this time, the index dictionary includes three index categories, namely, an anchor index, a partition index and a tag index, and the search intents respectively corresponding to the index categories are the anchor intention, the partition intention and the tag intention. The anchor index corresponds to a plurality of anchor index words, the partition index corresponds to a plurality of partition index words, and the tag index corresponds to a plurality of tag index words.
In this embodiment, the preset matching rule corresponding to the index category may be set according to the matching requirement in the actual application. The preset matching rules corresponding to different index categories may be the same or different. For example, when the index word dictionary includes a anchor index, a partition index, and a tag index, since the matching requirement of the partition index word and the tag index word is relatively high and the matching requirement of the anchor index is relatively low, the partition index and the tag index may correspond to a first preset matching rule having a relatively high matching requirement and the anchor index may correspond to a second preset matching rule that is different from the first preset matching rule and has a relatively low matching requirement.
In the index word dictionary, the live broadcast room corresponding to each index word is a popular live broadcast room under the corresponding intention of the index word, and the specific number can be set as required. It should be noted that index words with similar word senses may correspond to the same live broadcast room. For example, if a certain index word is a tag index word, and a specific corresponding tag is humorous, the live broadcast room corresponding to the index word may be a hot live broadcast room under the humorous tag. For example, the hot live room may be a fan-headed 20-bit live room.
As an embodiment, the step of obtaining a matching degree between the feature word and an index word in a preset index word dictionary includes: and acquiring an editing distance between the feature words and index words in a preset index word dictionary, and taking the editing distance as the matching degree between the feature words and the index words.
The editing distance is also called a Levenshtein distance, and refers to the minimum number of editing operations required for converting one character string into another character string. Permitted editing operations include replacing one character with another, inserting one character, and deleting one character. Generally, the smaller the edit distance, the greater the similarity of two character strings.
Of course, in other embodiments of the present invention, the matching degree between the feature word and the index word may also be obtained according to other algorithms, for example, the similarity between the feature word and the index word may also be calculated, and the similarity is used as the matching degree between the feature word and the index word.
Step S103, determining the index words corresponding to the search words according to the matching degree and the preset matching rules corresponding to each index category;
in this step, whether the matching degree between the feature word and the index word in the index word dictionary meets a preset condition is judged according to a preset matching rule corresponding to each index category, and the index word meeting the preset condition is used as the index word corresponding to the search word. Specifically, step S103 may include the following several embodiments:
in a first embodiment, the index dictionary includes a first index category. The first index categories include index categories corresponding to the same preset matching rule. For example, in one particular application scenario, the first index category may include the partition index and the tab index described above. At this time, as shown in fig. 2, step S103 may specifically include:
step S201, judging whether the matching degree between the feature words and first target index words meets a first preset condition, wherein the first target index words comprise index words corresponding to the first index category;
step S202, if the first preset condition is satisfied, taking the first target index word whose matching degree satisfies the first preset condition as the index word corresponding to the search word.
In this embodiment, the first index category corresponds to a plurality of index words, and the index words are all first target index words. For example, when the first index category includes a partition index and a tag index, the partition index word corresponding to the partition index and the tag index word corresponding to the tag index are both the index words corresponding to the first index category, that is, both the index words are the first target index words. That is, step S201 needs to determine whether the matching degree between the feature word and each first target index word satisfies a first preset condition.
As an implementation manner, the step of determining whether the matching degree between the feature word and the first target index word satisfies a first preset condition includes: and comparing the matching degree between the feature words and the first target index words with a preset target matching degree, and judging that a first preset condition is met when the matching degree is consistent with the target matching degree. And when the matching degree is inconsistent with the target matching degree, judging that the first preset condition is not met. The target matching degree can be set according to the actual application requirement. For example, in a case where the first index category includes the above-mentioned partition index and tag index, and the edit distance between the feature word and the first target index word is taken as the matching degree therebetween, the target matching degree may be set to 0, and at this time, it may be understood that when the feature value completely coincides with the first target index word, it is determined that the matching degree between the feature value and the first target index word satisfies the first preset condition.
Of course, in other embodiments of the present invention, the matching degree between the feature word and the first target index word may also be compared with a preset matching degree threshold, and when the matching degree exceeds the matching degree threshold, it is determined that the first preset condition is satisfied. And the first target index word with the matching degree not exceeding the threshold value of the matching degree does not meet the first preset condition. Wherein, the threshold value of the matching degree can be set according to the actual application requirement.
For example, the first target index word comprises index word A1… …, index word AM. Wherein M is an integer greater than 1.When the feature word and the index word A in the feature word1And index word A3When the matching degrees of the index words A all meet the first preset condition, the index words A are used for searching the index words A1And index word A3Are all used as index words corresponding to the search words.
It should be noted that, if the matching degrees between the feature words and all the first target index words do not satisfy the first preset condition, it is indicated that no first target index word corresponds to the feature value. In the case where the first index category includes a partition index and a tag index, it means that the feature word cannot be recognized as a partition intention or a tag intention.
In a second embodiment, the index dictionary includes a second index category. In this embodiment, the second index category is an index category different from the first index category. And the preset matching rule corresponding to the second index category is different from the matching rule corresponding to the first index category. For example, in one particular application scenario, the second index category may include the anchor index described above. At this time, as shown in fig. 3, step S103 may specifically include:
step S301, judging whether the matching degree between the feature words and second target index words meets a second preset condition, wherein the second target index words comprise index words corresponding to the second index category;
step S302, if the second preset condition is satisfied, taking a second target index word whose matching degree satisfies the second preset condition as the index word corresponding to the search word.
The second index category corresponds to a plurality of index words, and the index words are all second target index words. For example, when the second index category includes the anchor index described above, the anchor index words corresponding to the anchor index are all the index words corresponding to the second index category, that is, all the index words are the second target index words. That is, step S301 needs to determine whether the matching degree between the feature word and each second target index word satisfies a second preset condition.
As an embodiment, the step of determining whether the matching degree between the feature word and the second target index word satisfies a second preset condition may include:
obtaining a characteristic value according to the text length of the characteristic word, the text length of the second target index word, the matching degree between the characteristic word and the second target index word and a preset algorithm;
the text lengths of the feature words and the second target index words can be obtained through some text length calculation functions. For example, the text lengths of the feature word and the second target index word may be obtained by a length function.
Under the condition that the edit distance between the feature word and the first target index word is used as the matching degree of the feature word and the first target index word, the step of obtaining the feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word, and a preset algorithm may include: calculating a difference value between the text length of the second target index word and the text length of the feature word as a first difference value; calculating a difference value between the matching degree between the feature word and the first target index word and the first difference value as a second difference value; and taking the ratio of the second difference value to the text length of the feature word as the feature value.
Specifically, the preset algorithm may be the following formula:
wherein, length (q)i) Representation character word qiThe text length of (d); length (d)j) Represents a second target index word djThe text length of (d); dist (q)i,dj) Representation character word qiAnd a second target index word djThe edit distance between them, i.e. the degree of match; t represents a characteristic value. The characteristic word qiLength of text, second target index word djLength of text and feature word qiAnd a second target index word djThe characteristic word q can be obtained by inputting the above formula according to the edit distance between the charactersiAnd a second target index word djCorresponding special characterAnd (5) feature value.
After the characteristic value is obtained, it is further determined that the characteristic value is compared with a preset characteristic threshold value. If the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets the second preset condition; and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
Wherein, the characteristic threshold value can be set according to the specific application requirement. For example, when the second index category includes the anchor index described above, the second target index word is the anchor index word corresponding to the anchor index. The characteristic threshold may be set to 0.5, but of course, may be set to other values as desired.
Further, in order to simplify the calculation process, before calculating the feature values corresponding to the feature words and the second target index words, the text length of the feature words may be pre-determined, and for the feature words meeting the pre-determination conditions, the step of calculating the feature values may be performed, so that some unnecessary calculation processes may be avoided. That is to say, before the step of obtaining the feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word, and the preset algorithm is performed, a pre-determination step is further included. The pre-determining step may include: acquiring the text length of the feature words; judging whether the text length of the feature words is larger than a preset length threshold value or not; if the text length of the feature word is larger than a preset length threshold, executing the above-mentioned feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word and a preset algorithm; if the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets a second preset condition; and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
And for the feature words with the text length smaller than or equal to the preset length threshold, the feature value corresponding to the feature and the second target index word is not calculated. And judging that no second target index word corresponding to the characteristic value exists.
Specifically, the length threshold may be set as needed. For example, in the present embodiment, the length threshold may be set to 1.
It should be noted that, if the matching degrees between the feature words and all the second target index words do not satisfy the second preset condition, it is indicated that no second target index word corresponds to the feature value. In the case where the second index category includes a anchor index, it means that the feature word cannot be recognized as an anchor intention.
In a third embodiment, the index dictionary includes a first index category and a second index category. The second index category is an index category different from the first index category. And the preset matching rule corresponding to the second index category is different from the matching rule corresponding to the first index category. For example, in a specific application scenario, the first index category may include the partition index and the tab index described above, and the second index category may include the anchor index described above. At this time, as shown in fig. 4, step S103 may specifically include:
step S401, judging whether the matching degree between the feature words and first target index words meets a first preset condition, wherein the first target index words comprise index words corresponding to the first index category;
step S402, if the first preset condition is met, taking the first target index word of which the matching degree meets the first preset condition as the index word corresponding to the search word;
step S403, determining whether a matching degree between the feature word and a second target index word satisfies a second preset condition, where the second target index word includes an index word corresponding to the second index category;
step S404, if the second preset condition is satisfied, taking the second target index word whose matching degree satisfies the second preset condition as the index word corresponding to the search word.
Specific embodiments of steps S401 and S402 can refer to steps S201 and S202 in the first embodiment; the specific implementation of step S403 and step S404 may refer to step S301 and step S302 in the second implementation manner, which is not described herein again.
It should be noted that the sequence of steps shown in fig. 4 does not limit the smooth execution of steps S401 to S404 in this embodiment. Steps S401 to S404 may be executed according to the sequence shown in fig. 4, or may be executed according to another sequence, for example, step S401 and step S403 may be executed sequentially or substantially simultaneously, which is specifically set according to actual needs.
Alternatively, in another embodiment of the present invention, step S403 and step S404 may be executed again when the determination result in step S401 is that the matching degrees between the feature word and all the first target index words do not satisfy the first preset condition, and when the determination result in step S401 is that the matching degrees between the first target index word and the feature value satisfy the first preset condition, step S403 and step S404 may not be executed again.
It should be noted that, if the matching degrees between the feature word and all the first target index words do not satisfy the first preset condition, and the matching degrees between the feature word and all the second target index words do not satisfy the second preset condition, it is indicated that neither the first target index word nor the second target index word corresponds to the feature value. In the case where the first index category includes the above-described partition index and tag index, and the second index category includes the above-described anchor index, it means that the feature word cannot be recognized as neither the partition index nor the tag index, nor the anchor intention.
Of course, in addition to the above-mentioned several embodiments, in other embodiments of the present invention, the preset matching rule corresponding to each index category may also be a matching rule obtained in advance based on a rule template or a machine learning classification algorithm, so as to determine the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category.
And step S104, recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
After the steps S102 and S103 are performed on all the feature values acquired in step S101, an index word corresponding to the search word can be obtained. Because the index words correspond to the live broadcast rooms in advance, the live broadcast rooms corresponding to the index words can be recommended to the user according to the index words corresponding to the search words.
It should be further noted that, after the steps S102 and S103 are performed on all the feature values obtained in step S101, if the matching degrees between all the feature values and the index words in the index word dictionary do not meet the corresponding preset conditions, it indicates that no index word corresponds to a search word, and at this time, the recommendation result in step S104 is empty, that is, no live broadcast room meeting the intention of the search word is recommended to the user.
For example, in a specific embodiment, the index word dictionary includes a first index category and a second index category, the first index category includes a partition index and a tag index, the second index category includes a main index, and in this case, the first target index word includes a partition index word and a tag index word, and the second target index word includes a main index word. Assuming a search word Q, a characteristic word sequence { Q ] can be obtained by word segmentation1,q2,...qkH, feature word qiAnd an index word d in the index word dictionaryjAnd matching to obtain the index word corresponding to the search word Q. Wherein k, i and j are integers which are more than 1 or equal to 1. The method comprises the following specific steps:
(1) computing feature words qiAnd searching the index word d in the index dictionaryjEdit distance dist (q) betweeni,dj) As a feature word qiAnd an index word djThe degree of match between them.
(2) If the feature word qiThe edit distance from the partition index word or the tag index word is 0, then q isiIs identified as a partition intention or a tag intention, and an index word d with an edit distance of 0 is generatedjAs an index word corresponding to the search word Q. If there is no word q associated with the featureiIf the partition index word or the tag index word with the edit distance of 0 is edited, q cannot be expressediIdentified as a partition intent or a tag intent.
(3) Judging qiAnd whether the following relation is satisfied between the anchor index words, if so, q is setiIs identified as the anchor intention, and an anchor index word satisfying the following relationship is taken as an index word corresponding to the search word Q.
length(qi) Is greater than 1, and
for all characteristic values q in the characteristic word sequenceiBy doing the above operation, each feature value q can be obtainediCorresponding search intention, and corresponding index word dsi. The characteristic value q isiThe corresponding search term may be zero, may be one, or may be multiple. Thus, the search term Q may correspond to zero, one, or multiple index terms.
Suppose that the index word corresponding to the search word Q is d1And d50Wherein d is1For partitioning index words, d50If the search word is the anchor index word, the search word Q is identified as the partition intention and the anchor intention, and the partition index word d is recommended to the user1Corresponding live room and anchor index word d50A corresponding anchor's live room. Therefore, the search intention of the user can be quickly confirmed and targeted recommendation can be made according to the intention, and the algorithm is simple to implement and low in complexity.
In order to more clearly illustrate the live broadcast recommendation method provided by the present invention, a specific application scenario is taken as an example to illustrate the live broadcast recommendation method provided by an embodiment of the present invention.
Suppose that the search term of the user is "xx singing & & $", where "xx" denotes the name of a certain anchor. First, search words are preprocessed, where the special character "& & $" is deleted, and the processed search words become "xx singing". And then segmenting the processed search word to obtain two words of 'xx' and 'singing'. The two words are respectively matched with an index word dictionary, and the index word 'xx' can be matched with the main broadcasting index word 'von xx' through calculation, and the 'singing' can be matched with the label index word 'singing', so that the search intention is identified, and the search intention can be that the main broadcasting index word 'von xx' or other live broadcasting rooms with singing labels are wanted. The customer is then recommended a home play "von xx" and a popular live broadcast with singing labels.
The live broadcast room recommendation method provided by the embodiment of the invention comprises the steps of firstly obtaining the characteristic words of search words input by a user, then obtaining the matching degree between the characteristic words and the index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, each index word corresponds to a live broadcast room, then determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category, and further recommending the corresponding live broadcast room to the user according to the index words corresponding to the search words. Different index categories correspond to different search intentions, and the index words corresponding to the search words are determined by combining the search intentions and the matching degrees, so that the reliability of the recommendation result of the live broadcast room is improved, namely, the relevance between the recommended live broadcast room and the real intention of the user is improved, and the user experience is improved.
Referring to fig. 5, a second embodiment of the present invention provides a live broadcast room recommendation apparatus. As shown in fig. 5, the live broadcast room recommendation apparatus includes: a feature word obtaining module 501, a matching degree obtaining module 502, an index word determining module 503, and a recommending module 504.
The feature word obtaining module 501 is configured to obtain a feature word of a search word input by a user.
A matching degree obtaining module 502, configured to obtain a matching degree between the feature word and an index word in a preset index word dictionary, where the index word dictionary includes one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast.
An index word determining module 503, configured to determine the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category.
A recommending module 504, configured to recommend the corresponding live broadcast room to the user according to the index word corresponding to the search word.
As an alternative embodiment, the index dictionary includes a first index category, and the index word determining module 503 includes:
the first judgment sub-module is used for judging whether the matching degree between the feature words and first target index words meets a first preset condition or not, wherein the first target index words comprise index words corresponding to the first index category;
and the first determining submodule is used for taking the first target index word of which the matching degree meets the first preset condition as the index word corresponding to the search word if the first preset condition is met.
As an alternative embodiment, the index dictionary includes a second index category, and the index word determining module 503 includes:
the second judgment submodule is used for judging whether the matching degree between the feature words and second target index words meets a second preset condition or not, wherein the second target index words comprise index words corresponding to the second index category;
and the second determining submodule is used for taking the second target index word of which the matching degree meets the second preset condition as the index word corresponding to the search word if the second preset condition is met.
As an alternative embodiment, the index dictionary includes a first index category and a second index category. As shown in fig. 6, the index word determining module 503 includes:
a first determining sub-module 601, configured to determine whether a matching degree between the feature word and a first target index word meets a first preset condition, where the first target index word includes an index word corresponding to the first index category;
a first determining sub-module 602, configured to, if the first preset condition is met, take a first target index word whose matching degree meets the first preset condition as the index word corresponding to the search word;
a second determining sub-module 603, configured to determine whether a matching degree between the feature word and a second target index word meets a second preset condition, where the second target index word includes an index word corresponding to the second index category;
a second determining sub-module 604, configured to, if the second preset condition is met, take a second target index word whose matching degree meets the second preset condition as the index word corresponding to the search word.
As an alternative embodiment, the second determining sub-module 603 is specifically configured to: obtaining a characteristic value according to the text length of the characteristic word, the text length of the second target index word, the matching degree between the characteristic word and the second target index word and a preset algorithm; if the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets a second preset condition; and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
As an alternative embodiment, the second determining sub-module 603 is specifically configured to: acquiring the text length of the feature words; judging whether the text length of the feature words is larger than a preset length threshold value or not; if the text length of the feature word is larger than the length threshold, obtaining a feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word and a preset algorithm; if the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets a second preset condition; and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
As an optional embodiment, the matching degree obtaining module 502 is specifically configured to: and acquiring an editing distance between the feature words and index words in a preset index word dictionary, and taking the editing distance as the matching degree between the feature words and the index words.
As an optional embodiment, the feature word obtaining module 501 is specifically configured to: acquiring a search word input by a user; and performing word segmentation processing on the search word to obtain one or more search words, and taking the one or more search words as the characteristic word.
It should be noted that, the implementation principle and the generated technical effect of the live broadcast recommendation device provided by the embodiment of the present invention are the same as those of the foregoing method embodiments, and for a brief description, reference may be made to corresponding contents in the foregoing method embodiments for a part not mentioned in the embodiment of the device.
A third embodiment of the invention provides an electronic device comprising a processor and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to:
acquiring characteristic words of search words input by a user;
acquiring the matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises one or more index categories, each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast;
determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category;
and recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word.
Specifically, the electronic device may be a server, or may also be a user terminal. The user terminal may include a pc (personal computer), a tablet computer, a mobile phone, a notebook computer, a smart television, and other terminal devices.
A fourth embodiment of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the first embodiment described above. The functional unit integrated with the live broadcast recommendation device in the second embodiment of the present invention may be stored in a computer readable storage medium if it is implemented in the form of a software functional unit and sold or used as an independent product. Based on such understanding, all or part of the flow of the live broadcast recommendation method according to the first embodiment may be implemented by a computer program, which may be stored in a computer-readable storage medium and may be executed by a processor to implement the steps of the above method embodiments. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, the present invention is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components of a gateway, proxy server, system according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
Claims (10)
1. A live broadcast room recommendation method is characterized by comprising the following steps:
acquiring characteristic words of search words input by a user;
obtaining a matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises a plurality of index categories, each index category is used for reflecting a live broadcast search intention of a user, and the index categories comprise: the index system comprises a main broadcast index, a partition index and a label index, wherein each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast room;
determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category;
if an index word corresponding to the search word exists in the index word dictionary, recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word, and if the index word corresponding to the search word does not exist in the index word dictionary, recommending a result to be null, wherein the recommending the corresponding live broadcast room to the user comprises: recommending the live broadcast room corresponding to the index word to the user.
2. The method of claim 1, wherein the index dictionary comprises a first index category, and the step of determining the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category comprises:
judging whether the matching degree between the feature words and first target index words meets a first preset condition or not, wherein the first target index words comprise index words corresponding to the first index category;
and if the first preset condition is met, taking the first target index word with the matching degree meeting the first preset condition as the index word corresponding to the search word.
3. The method of claim 1, wherein the index word dictionary includes a second index category, and the step of determining the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category comprises:
judging whether the matching degree between the feature words and second target index words meets a second preset condition or not, wherein the second target index words comprise index words corresponding to the second index category;
and if the second preset condition is met, taking the second target index word with the matching degree meeting the second preset condition as the index word corresponding to the search word.
4. The method according to claim 3, wherein the step of determining whether the matching degree between the feature word and the second target index word satisfies a second preset condition comprises:
obtaining a characteristic value according to the text length of the characteristic word, the text length of the second target index word, the matching degree between the characteristic word and the second target index word and a preset algorithm;
if the characteristic value is smaller than a preset characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word meets a second preset condition;
and if the characteristic value is not smaller than the characteristic threshold value, judging that the matching degree between the characteristic word and a second target index word does not meet the second preset condition.
5. The method as claimed in claim 4, wherein before obtaining the feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word, and a preset algorithm, the method further comprises:
acquiring the text length of the feature words;
judging whether the text length of the feature words is larger than a preset length threshold value or not;
and if the text length of the feature word is larger than the length threshold, executing the step of obtaining a feature value according to the text length of the feature word, the text length of the second target index word, the matching degree between the feature word and the second target index word and a preset algorithm.
6. The method of claim 1, wherein the step of obtaining the matching degree between the feature words and the index words in a preset index word dictionary comprises:
and acquiring an editing distance between the feature words and index words in a preset index word dictionary, and taking the editing distance as the matching degree between the feature words and the index words.
7. The method of claim 1, wherein the step of obtaining the feature words of the search words input by the user comprises:
acquiring a search word input by a user;
and performing word segmentation processing on the search word to obtain one or more search words, and taking the one or more search words as the characteristic word.
8. A live room recommendation apparatus, characterized in that the apparatus comprises:
the characteristic word acquisition module is used for acquiring the characteristic words of the search words input by the user;
a matching degree obtaining module, configured to obtain a matching degree between the feature word and an index word in a preset index word dictionary, where the index word dictionary includes multiple index categories, each index category is used to reflect a search intention of a user in a live broadcast, and the multiple index categories include: the index system comprises a main broadcast index, a partition index and a label index, wherein each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast room;
an index word determining module, configured to determine the index word corresponding to the search word according to the matching degree and the preset matching rule corresponding to each index category;
a recommending module, configured to recommend the live broadcast to the user according to the index word corresponding to the search word if the index word corresponding to the search word exists in the index word dictionary, and recommend a result as null if the index word corresponding to the search word does not exist in the index word dictionary, where recommending the corresponding live broadcast to the user includes: recommending the live broadcast room corresponding to the index word to the user.
9. An electronic device comprising a processor and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to:
acquiring characteristic words of search words input by a user;
obtaining a matching degree between the feature words and index words in a preset index word dictionary, wherein the index word dictionary comprises a plurality of index categories, each index category is used for reflecting a live broadcast search intention of a user, and the index categories comprise: the index system comprises a main broadcast index, a partition index and a label index, wherein each index category corresponds to a preset matching rule and a plurality of index words, and each index word corresponds to a live broadcast room;
determining the index words corresponding to the search words according to the matching degree and the preset matching rule corresponding to each index category;
if an index word corresponding to the search word exists in the index word dictionary, recommending the corresponding live broadcast room to the user according to the index word corresponding to the search word, and if the index word corresponding to the search word does not exist in the index word dictionary, recommending a result to be null, wherein the recommending the corresponding live broadcast room to the user comprises: recommending the live broadcast room corresponding to the index word to the user.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810029064.2A CN108256044B (en) | 2018-01-12 | 2018-01-12 | Recommended method, device and electronic equipment for live broadcast room |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810029064.2A CN108256044B (en) | 2018-01-12 | 2018-01-12 | Recommended method, device and electronic equipment for live broadcast room |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108256044A CN108256044A (en) | 2018-07-06 |
| CN108256044B true CN108256044B (en) | 2021-04-27 |
Family
ID=62727150
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810029064.2A Expired - Fee Related CN108256044B (en) | 2018-01-12 | 2018-01-12 | Recommended method, device and electronic equipment for live broadcast room |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108256044B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110896488B (en) * | 2018-08-23 | 2022-01-04 | 武汉斗鱼网络科技有限公司 | Recommendation method for live broadcast room and related equipment |
| CN111104583B (en) * | 2018-10-10 | 2024-01-05 | 河南星易网络科技有限公司 | Live broadcast room recommendation method, storage medium, electronic equipment and system |
| CN109547863B (en) * | 2018-10-22 | 2021-06-15 | 武汉斗鱼网络科技有限公司 | Label marking method, label marking device, server and storage medium |
| CN110430476B (en) * | 2019-08-05 | 2021-12-28 | 广州方硅信息技术有限公司 | Live broadcast room searching method, system, computer equipment and storage medium |
| CN110941765A (en) * | 2019-12-04 | 2020-03-31 | 青梧桐有限责任公司 | Search intention identification method, information search method and device and electronic equipment |
| CN112040257A (en) * | 2020-08-18 | 2020-12-04 | 北京达佳互联信息技术有限公司 | Method and device for pushing information in live broadcast room |
| CN113450803B (en) * | 2021-06-09 | 2024-03-19 | 上海明略人工智能(集团)有限公司 | Conference recording transfer method, system, computer device and readable storage medium |
| CN114025176A (en) * | 2021-08-24 | 2022-02-08 | 广州方硅信息技术有限公司 | Anchor recommendation method and device, electronic equipment and storage medium |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8645125B2 (en) * | 2010-03-30 | 2014-02-04 | Evri, Inc. | NLP-based systems and methods for providing quotations |
| CN102033947B (en) * | 2010-12-22 | 2013-01-16 | 百度在线网络技术（北京）有限公司 | Region recognizing device and method based on retrieval word |
| CN103914479B (en) * | 2013-01-06 | 2017-12-01 | 北京金山安全软件有限公司 | Resource request matching method and device |
| US20150088649A1 (en) * | 2013-09-22 | 2015-03-26 | Acxiom Corporation | Quick Audience Search and Recommendation Apparatus and Method |
| CN104408191B (en) * | 2014-12-15 | 2017-11-21 | 北京国双科技有限公司 | The acquisition methods and device of the association keyword of keyword |
| CN105956148A (en) * | 2016-05-12 | 2016-09-21 | 北京奇艺世纪科技有限公司 | Resource information recommendation method and apparatus |
| CN107092642A (en) * | 2017-03-06 | 2017-08-25 | 广州神马移动信息科技有限公司 | A kind of information search method, equipment, client device and server |
| CN106971000B (en) * | 2017-04-12 | 2020-04-28 | 北京焦点新干线信息技术有限公司 | A search method and device |
| CN107273537A (en) * | 2017-06-30 | 2017-10-20 | 深圳创维数字技术有限公司 | One kind search words recommending method, set top box and storage medium |
-
2018
- 2018-01-12 CN CN201810029064.2A patent/CN108256044B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN108256044A (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108256044B (en) | Recommended method, device and electronic equipment for live broadcast room | |
| CN114549874B (en) | Training method of multi-target image-text matching model, image-text retrieval method and device | |
| CN109885770B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN109325146B (en) | Video recommendation method and device, storage medium and server | |
| CN112182348B (en) | Semantic matching determination method, device, electronic equipment, computer-readable medium | |
| CN109348262B (en) | Calculation method, device, equipment and storage medium for anchor similarity | |
| CN110019948B (en) | Method and apparatus for outputting information | |
| CN115374369A (en) | News diversity recommendation method and device based on graph neural network | |
| CN109190116B (en) | Semantic analysis method, system, electronic device and storage medium | |
| KR102779318B1 (en) | Method, computer device, and computer program to recommend similar product based on keyword | |
| CN112989177B (en) | Information processing method, information processing device, electronic equipment and computer storage medium | |
| CN110610001B (en) | Short text integrity recognition method, device, storage medium and computer equipment | |
| CN109815312B (en) | Document query method and device, computing equipment and computer storage medium | |
| CN112287655A (en) | Matched text duplicate removal method and device, and electronic equipment | |
| WO2016101737A1 (en) | Search query method and apparatus | |
| CN113849688B (en) | Resource processing method, resource processing device, electronic device and storage medium | |
| CN112883218A (en) | Image-text combined representation searching method, system, server and storage medium | |
| CN112581954B (en) | High-matching voice interaction method and intelligent device | |
| CN111753050B (en) | Topic map-based comment generation | |
| CN116028704A (en) | Recommended method and device, equipment, storage medium | |
| CN114048376A (en) | Advertisement service information mining method and device, electronic equipment and storage medium | |
| CN112288548B (en) | Method, device, medium and electronic device for extracting key information of target object | |
| KR20210056668A (en) | Method and system for providing related kewords with respect to keyword input by user for in order to deriving idea | |
| CN112183069A (en) | Keyword construction method and system based on historical keyword release data | |
| CN116738973B (en) | A search intention recognition method, a prediction model construction method and an electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210427 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |