CN110309410B - Information recommendation method, platform and computer readable storage medium - Google Patents

Information recommendation method, platform and computer readable storage medium Download PDF

Info

Publication number
CN110309410B
CN110309410B CN201810214568.1A CN201810214568A CN110309410B CN 110309410 B CN110309410 B CN 110309410B CN 201810214568 A CN201810214568 A CN 201810214568A CN 110309410 B CN110309410 B CN 110309410B
Authority
CN
China
Prior art keywords
information
recommendation
similarity
browsing data
recommendation list
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810214568.1A
Other languages
Chinese (zh)
Other versions
CN110309410A (en
Inventor
李小文
李晟
郭洪波
王艳彬
杨东
雷敏
邢荣荣
李关乐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Communications Group Co Ltd
China Mobile Chengdu ICT Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Mobile Communications Group Co Ltd filed Critical China Mobile Communications Group Co Ltd
Priority to CN201810214568.1A priority Critical patent/CN110309410B/en
Publication of CN110309410A publication Critical patent/CN110309410A/en
Application granted granted Critical
Publication of CN110309410B publication Critical patent/CN110309410B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation

Abstract

The embodiment of the invention discloses an information recommendation method, a platform and a computer readable storage medium, wherein the information recommendation method comprises the following steps: acquiring first browsing data corresponding to a recommended object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.

Description

Information recommendation method, platform and computer readable storage medium
Technical Field
The present invention relates to the field of information services, and in particular, to an information recommendation method, a platform, and a computer-readable storage medium.
Background
Currently, when an information recommendation platform recommends information for a user, available recommendation methods include: content-based recommendations, collaborative filtering recommendations, association-based recommendations, utility-based recommendations, knowledge-based recommendations, and combination recommendations. Among them, methods of content-based recommendation, collaborative filtering recommendation, and combined recommendation are most commonly used. Specifically, the content-based recommendation is to find the relevance of the information content according to the metadata of the information content, and then recommend similar information to the user based on the browsing data of the user; collaborative filtering recommendation mainly comprises three modes, namely collaborative filtering recommendation based on a user, collaborative filtering recommendation based on a project and collaborative filtering recommendation based on a model; the combined recommendation is to recommend the information in different combinations, including weighting, transformation, mixing, stacking, and expansion.
In the existing information recommendation mode, content-based recommendation needs to be combined with browsing data of users for information recommendation, so that for users without browsing data, the method has the problems of sparseness and recommendation failure; the extensibility of collaborative filtering recommendation is poor, and the recommendation quality depends on browsing data of a user, so the method cannot recommend information to the user without the browsing data; although the combined recommendation method can overcome some problems of content-based recommendation and collaborative filtering recommendation, the combined recommendation method is mostly implemented in combination with specific scenes and specific applications, and needs to refer to browsing data of users for recommendation, so that a problem that information recommendation cannot be performed for users without browsing data still exists, that is, the existing information recommendation technology cannot perform information recommendation for any type of users, that is, the existing technology cannot perform personalized information recommendation for users.
Disclosure of Invention
In order to solve the above technical problems, embodiments of the present invention are intended to provide an information recommendation method, a platform, and a computer-readable storage medium, which can solve the problem that information recommendation cannot be performed on a recommendation object without browsing data, so that personalized information recommendation can be performed on any type of recommendation object.
In order to achieve the above purpose, the technical solution of the embodiment of the present invention is realized as follows:
the embodiment of the invention provides an information recommendation method, which comprises the following steps:
acquiring first browsing data corresponding to a recommended object;
when the first browsing data exists in the preset source information, acquiring a first recommendation list according to the information attribute and the information content corresponding to the first browsing data;
when the first browsing data does not exist in the preset source information, acquiring a second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object;
and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
In the above scheme, when the preset source information has the first browsing data, acquiring a first recommendation list according to the information attribute and the information content corresponding to the first browsing data, including:
determining attribute similarity and content similarity corresponding to the first browsing data and preset source information according to the information attribute, the information content and the preset source information;
and determining the first recommendation list according to the attribute similarity and the content similarity.
In the above scheme, the determining attribute similarity corresponding to the first browsing data and the preset source information according to the information attribute and the preset source information includes:
acquiring a pre-stored attribute corresponding to the preset source information;
and inputting the information attributes and the pre-stored attributes into a preset attribute similarity calculation model to obtain the attribute similarity.
In the above scheme, the determining content similarity corresponding to the first browsing data and the preset source information according to the information content and the preset source information includes:
acquiring prestored contents corresponding to the preset source information;
determining a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content;
and inputting the first keyword weight and the second keyword weight into a preset content similarity calculation model to obtain the content similarity.
In the foregoing solution, the determining the first recommendation list according to the attribute similarity and the content similarity includes:
inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information;
determining a first online time corresponding to the preset source information;
and determining the first recommendation list according to the information similarity and the first online time.
In the foregoing scheme, when the first browsing data does not exist in the preset source information, obtaining a second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object includes:
acquiring a target object according to the first browsing data, the attribute information and a pre-stored object identifier;
acquiring second browsing data corresponding to the target object;
and determining the second recommendation list according to the second browsing data.
In the foregoing scheme, the obtaining a target object according to the first browsing data, the attribute information, and a pre-stored object identifier includes:
acquiring identification information corresponding to the recommended object according to the first browsing data and the attribute information;
inputting the identification information and the pre-stored object identification into a preset object similarity calculation model to obtain object similarity;
and determining the target object according to the object similarity.
In the foregoing solution, the determining the second recommendation list according to the second browsing data includes:
determining second online time corresponding to the second browsing data;
and determining the second recommendation list according to the second browsing data and the second online time.
In the foregoing solution, after the first browsing data corresponding to the recommended object is obtained, the method further includes:
acquiring a third recommendation list according to a preset heat type; the preset popularity type is used for recommending the information according to the popularity of the information.
In the foregoing scheme, the obtaining a third recommendation list according to a preset heat type includes:
acquiring the access amount corresponding to the preset source information;
inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to the preset source information;
and acquiring the third recommendation list according to the heat parameter.
In the above solution, the recommending information to the recommendation object according to the first recommendation list or the second recommendation list includes:
according to the first recommendation list and the third recommendation list, recommending information to the recommendation object; alternatively, the first and second electrodes may be,
and recommending information to the recommended object according to the second recommendation list and the third recommendation list.
In the above solution, the recommending information to the recommendation object according to the first recommendation list and the third recommendation list includes:
determining a first information quantity corresponding to the first recommendation list;
if the first information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the first recommendation list so as to recommend information to the recommendation object;
and if the first information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the first recommendation list.
In the above solution, the recommending information to the recommendation object according to the second recommendation list and the third recommendation list includes:
determining a second information quantity corresponding to the second recommendation list;
if the second information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the second recommendation list so as to recommend information to the recommendation object;
and if the second information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the second recommendation list.
The embodiment of the invention provides an information recommendation platform, which comprises: an acquisition unit and a recommendation unit,
the acquisition unit is used for acquiring first browsing data corresponding to the recommended object; when the preset source information has the first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the first browsing data does not exist in the preset source information, acquiring a second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object;
and the recommending unit is used for recommending information to the recommending object according to the first recommending list or the second recommending list.
In the above scheme, the obtaining unit is further configured to obtain a third recommendation list according to a preset heat type after obtaining the first browsing data corresponding to the recommendation object; the preset heat type is used for recommending information according to the heat of the information;
the recommending unit is further used for recommending information to the recommending object according to the first recommending list and the third recommending list; or recommending information to the recommendation object according to the second recommendation list and the third recommendation list.
The embodiment of the invention provides an information recommendation platform, which comprises: a processor, a memory, and a communication bus; the communication bus is used for realizing connection communication between the processor and the memory; the processor is used for executing the data transmission program stored in the memory so as to realize the following steps:
acquiring first browsing data corresponding to a recommended object;
when the first browsing data exists in the preset source information, acquiring a first recommendation list according to the information attribute and the information content corresponding to the first browsing data;
when the first browsing data does not exist in the preset source information, acquiring a second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object;
and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
In the above scheme, the processor is specifically configured to determine attribute similarity and content similarity corresponding to the first browsing data and preset source information according to the information attribute, the information content, and the preset source information; determining the first recommendation list according to the attribute similarity and the content similarity; and acquiring the pre-stored attribute corresponding to the preset source information; inputting the information attribute and the pre-stored attribute into a preset attribute similarity calculation model to obtain the attribute similarity; and acquiring prestored contents corresponding to the preset source information; determining a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content; inputting the first keyword weight and the second keyword weight into a preset content similarity calculation model to obtain the content similarity; inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information; determining a first online time corresponding to the preset source information; and determining the first recommendation list according to the information similarity and the first online time.
In the foregoing scheme, the processor is further specifically configured to obtain a target object according to the first browsing data, the attribute information, and a pre-stored object identifier; acquiring second browsing data corresponding to the target object; determining the second recommendation list according to the second browsing data; acquiring identification information corresponding to the recommended object according to the first browsing data and the attribute information; inputting the identification information and the pre-stored object identification into a preset object similarity calculation model to obtain object similarity; and determining the target object according to the object similarity; determining a second online time corresponding to the second browsing data; and determining the second recommendation list according to the second browsing data and the second online time.
In the above scheme, the processor is further configured to obtain a third recommendation list according to a preset heat type after obtaining first browsing data corresponding to the recommendation object; the preset heat type is used for recommending information according to the heat of the information;
the processor is further specifically configured to obtain an access amount corresponding to the preset source information; inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to the preset source information; acquiring the third recommendation list according to the heat parameter; and recommending information to the recommended object according to the first recommendation list and the third recommendation list; or, according to the second recommendation list and the third recommendation list, recommending information to the recommendation object; determining a first information quantity corresponding to the first recommendation list; if the first information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the first recommendation list so as to recommend information to the recommendation object; if the first information quantity is larger than or equal to the preset quantity threshold value, information recommendation is directly carried out on the recommendation object according to the first recommendation list; determining a second information quantity corresponding to the second recommendation list; if the second information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the second recommendation list so as to recommend information to the recommendation object; and if the second information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the second recommendation list.
An embodiment of the present invention provides a computer-readable storage medium, which stores one or more programs that are executable by one or more processors to implement the information recommendation method.
Therefore, in the technical scheme of the embodiment of the invention, the first browsing data corresponding to the recommended object is obtained; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, according to the information recommendation method, the information recommendation platform and the computer-readable storage medium provided by the embodiment of the invention, whether the first browsing data corresponding to the recommended object belongs to the preset source information or not can be determined, then the first recommendation list is obtained according to the information attribute and the information content corresponding to the first browsing data, or the second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommended object, and finally the personalized information recommendation is performed on the recommended object according to the first recommendation list or the second recommendation list, so that the problem that the information recommendation cannot be performed on the recommended object without the browsing data can be solved, and the personalized information recommendation can be performed on any type of recommended object.
Drawings
Fig. 1 is a schematic view illustrating a first implementation flow of an information recommendation method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a second implementation flow of an information recommendation method according to an embodiment of the present invention;
FIG. 3 is a third schematic flow chart illustrating an implementation of an information recommendation method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a fourth implementation flow of an information recommendation method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention;
fig. 6 is a schematic view illustrating a sixth implementation flow of an information recommendation method according to an embodiment of the present invention;
fig. 7 is a seventh schematic flow chart illustrating an implementation of an information recommendation method according to an embodiment of the present invention;
fig. 8 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention;
fig. 9 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention;
fig. 10 is a schematic flow chart illustrating an implementation of an information recommendation method according to an embodiment of the present invention;
fig. 11 is an eleventh schematic flow chart illustrating an implementation of an information recommendation method according to an embodiment of the present invention;
fig. 12 is a flowchart illustrating a twelfth implementation process of an information recommendation method according to an embodiment of the present invention;
FIG. 13 is a first schematic structural diagram of an information recommendation platform according to an embodiment of the present invention;
fig. 14 is a schematic structural diagram of an information recommendation platform according to an embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Example one
Fig. 1 is a schematic flow chart illustrating an implementation process of an information recommendation method according to an embodiment of the present invention, as shown in fig. 1, in an embodiment of the present invention, a method for information recommendation by an information recommendation platform mainly includes the following steps:
step 101, obtaining first browsing data corresponding to a recommended object.
In an embodiment of the invention, the information recommendation platform may first obtain first browsing data corresponding to a recommended object. The first browsing data may include at least one piece of information browsed by the recommended object.
It should be noted that, in the embodiment of the present invention, the information recommendation platform may be used for a news information platform that performs personalized recommendation for different recommendation objects.
Further, in an embodiment of the present invention, the first browsing data may include a plurality of different types of information browsed by the recommended object.
Step 102, when the preset source information has the first browsing data, acquiring a first recommendation list according to the information attribute and the information content corresponding to the first browsing data.
In an embodiment of the invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, if the preset source information has the first browsing data, the information recommendation platform may obtain a first recommendation list according to an information attribute and an information content corresponding to the first browsing data.
It should be noted that, in an embodiment of the present invention, after the information recommendation platform obtains the first browsing data, the information recommendation platform may compare the preset source information with the first browsing data, and if the first browsing data exists in the preset original information, it is determined that the recommendation object browses the preset source information, the information recommendation platform may further obtain the first recommendation list.
Further, in an embodiment of the present invention, if the first browsing data exists in the preset source information, the information recommendation platform may further obtain the first recommendation list according to information attributes and information contents corresponding to the first browsing data.
It should be noted that, in the embodiment of the present invention, the predetermined source information is a type of information in the information recommendation platform.
Step 103, when the first browsing data does not exist in the preset source information, a second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommendation object.
In an embodiment of the invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, if the preset source information does not have the first browsing data, a second recommendation list may be obtained according to the first browsing data and the attribute information corresponding to the recommendation object.
It should be noted that, in an embodiment of the present invention, after the information recommendation platform obtains the first browsing data, the information recommendation platform may compare the preset source information with the first browsing data, and if the first browsing data does not exist in the preset original information, it is determined that the recommendation object has not browsed the preset source information, the information recommendation platform may further obtain the second recommendation list.
Further, in an embodiment of the present invention, if the first browsing data does not exist in the preset source information, the information recommendation platform needs to further obtain the second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object.
And 104, recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
In an embodiment of the present invention, after the information recommendation platform obtains a first recommendation list according to the information attribute and the information content corresponding to the first browsing data, or after the information recommendation platform obtains a second recommendation list according to the attribute information corresponding to the recommendation object and the first browsing data, the information recommendation platform may further recommend information to the recommendation object according to the first recommendation list or the second recommendation list.
Further, in the embodiment of the present invention, when the recommendation object browses the preset source information, the information recommendation platform may recommend information to the recommendation object according to the first recommendation list; accordingly, when the recommendation object does not browse the preset source information, the information recommendation platform can recommend information to the recommendation object according to the second recommendation list. Therefore, personalized information recommendation can be effectively carried out according to different recommendation objects.
The information recommendation method provided by the embodiment of the invention obtains first browsing data corresponding to a recommendation object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, according to the information recommendation method provided by the embodiment of the invention, whether the first browsing data corresponding to the recommended object belongs to the preset source information or not can be determined, then the first recommendation list is obtained according to the information attribute and the information content corresponding to the first browsing data, or the second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommended object, and finally the recommended object is subjected to personalized information recommendation according to the first recommendation list or the second recommendation list, so that the problem that the recommended object without browsing data cannot be subjected to information recommendation can be solved, and further the personalized information recommendation can be performed on any type of recommended object.
Example two
Based on the first embodiment, fig. 2 is a schematic view of an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 2, in an embodiment of the present invention, when first browsing data exists in preset source information, a method for obtaining a first recommendation list according to information attributes and information contents corresponding to the first browsing data may include the following steps:
step 102a, determining attribute similarity and content similarity corresponding to the first browsing data and the preset source information according to the information attribute, the information content and the preset source information.
In an embodiment of the invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, if a preset source information has the first browsing data, the information recommendation platform may determine an attribute similarity and a content similarity corresponding to the first browsing data and the preset source information according to the information attribute, the information content, and a preset source information stored in advance.
It should be noted that, in an embodiment of the present invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, if the preset source information has the first browsing data, the information attribute and the information content corresponding to the first browsing data may be determined first.
It should be noted that, in an embodiment of the present invention, the information attribute is at least one attribute feature corresponding to at least one piece of information in the first browsing data.
Further, in an embodiment of the present invention, the attribute feature may be represented by an attribute vector. Specifically, in the embodiment of the present invention, for the information a, the information recommendation platform may extract basic attributes of the information a, for example, 4 of the type, topic, originality, and popularity of the information a is extracted as the basic attributes, and then the information recommendation platform may perform dummy variable on the basic attributes of the information a to obtain an attribute vector Ra of the information a, where the attribute vector Ra may represent attribute characteristics of the information a.
It should be noted that, in an embodiment of the present invention, the information content is at least one content feature corresponding to at least one piece of information in the first browsing data.
Further, in an embodiment of the present invention, the information recommendation platform may determine a keyword corresponding to the information in the first history browsing as a content feature corresponding to the information. Specifically, in the embodiment of the present invention, for the information a, the information recommendation platform may perform word segmentation and part-of-speech tagging on the title and the text of the information a, retain verbs, nouns and vernouns, directly use words appearing in the title as keywords, and set the weight as 1; and obtaining the weight of each word by using the word frequency and the word coverage of the words appearing in the text, then arranging the weights in a descending order, and taking at least one keyword with a larger weight value as the content characteristic corresponding to the information A.
Further, in an embodiment of the present invention, the attribute similarity is a similarity of information attributes between information in the first browsing data and other information in the predetermined source information; accordingly, the content similarity is the similarity of the information content between the information in the first browsing data and other information in the predetermined source information.
Specifically, in an embodiment of the present invention, the information recommendation platform may determine the attribute similarity corresponding to the first browsing data and the preset source information according to the information attribute and the preset source information; the information recommendation platform can determine the content similarity corresponding to the first browsing data and the preset source information according to the information content and the preset source information.
And step 102b, determining a first recommendation list according to the attribute similarity and the content similarity.
In an embodiment of the present invention, after the information recommendation platform determines the attribute similarity and the content similarity corresponding to the first browsing data and the pre-stored pre-set source information according to the information attribute, the information content, and the pre-stored pre-set source information, the information recommendation platform may determine the first recommendation list according to the attribute similarity and the content similarity.
Further, in an embodiment of the present invention, the information recommendation platform may determine at least one piece of information with a higher similarity to the first browsing data from the preset source information according to the attribute similarity and the content similarity when obtaining the attribute similarity and the content similarity, respectively, so as to use the at least one piece of information as the first recommendation list.
According to the description, through the steps 102a-102b, the information recommendation platform determines the attribute similarity and the content similarity corresponding to the first browsing data and the preset source information according to the information attribute, the information content and the preset source information; and determining a first recommendation list according to the attribute similarity and the content similarity. Therefore, according to the information recommendation method provided by the embodiment of the invention, whether the first browsing data corresponding to the recommended object belongs to the preset source information or not can be determined, then the first recommendation list is obtained according to the information attribute and the information content corresponding to the first browsing data, or the second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommended object, and finally the recommended object is subjected to personalized information recommendation according to the first recommendation list or the second recommendation list, so that the problem that the recommended object without browsing data cannot be subjected to information recommendation can be solved, and further the personalized information recommendation can be performed on any type of recommended object.
EXAMPLE III
Based on the second embodiment, fig. 3 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 3, in an embodiment of the present invention, a method for determining attribute similarity between first browsing data and preset source information by an information recommendation platform according to information attributes and the preset source information may include the following steps:
step 201, obtaining a pre-stored attribute corresponding to the preset source information.
In an embodiment of the invention, after determining the information attribute corresponding to the first browsing data, the information recommendation platform may first obtain a pre-stored attribute corresponding to the predetermined source information.
It should be noted that, in an embodiment of the present invention, the pre-stored attribute is at least one attribute feature corresponding to at least one of the preset source information.
Further, in the embodiment of the present invention, the attribute feature may also be represented by an attribute vector. Specifically, in the embodiment of the present invention, for the information B, the information recommendation platform may extract basic attributes of the information B, for example, 4 of the type, topic, originality, and hotness of the information B is extracted as basic attributes, and then the information recommendation platform may perform dummy variable on the basic attributes of the information B to obtain an attribute vector Rb of the information B, where the attribute vector Rb may represent attribute features of the information B.
Step 202, inputting the information attributes and the pre-stored attributes into a preset attribute similarity calculation model to obtain the attribute similarity.
In an embodiment of the present invention, after the information recommendation platform obtains the pre-stored attribute corresponding to the preset source information, the information attribute and the pre-stored attribute may be input into a preset attribute similarity calculation maze, and then the attribute similarity may be obtained.
Further, in an embodiment of the invention, the predetermined attribute similarity calculation model is used for calculating a similarity between different information according to different attribute characteristics corresponding to the different information. Specifically, in the embodiment of the present invention, the information recommendation platform may calculate and obtain the attribute similarity sim (i, j) between the information i and the information j according to the formula (1).
Figure BDA0001598215710000141
Wherein R isiAnd RjRespectively representing the vector attributes of the information i and the information j.
Fig. 4 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 4, in an embodiment of the present invention, the method for determining content similarity corresponding to first browsing data and preset source information according to information content and the preset source information may include the following steps:
step 301, obtaining the pre-stored content corresponding to the preset source information.
In an embodiment of the invention, after determining the information content corresponding to the first browsing data, the information recommendation platform may first obtain a pre-stored content corresponding to the predetermined source information.
It should be noted that, in the embodiment of the present invention, the pre-stored content is at least one content feature corresponding to at least one of the preset source information.
Further, in an embodiment of the present invention, the information recommendation platform may determine a keyword corresponding to information in the preset source information as a content feature corresponding to the information. Specifically, in the embodiment of the present invention, for the information B, the information recommendation platform may perform word segmentation and part-of-speech tagging on the title and the text of the information B, retain verbs, nouns and vernouns, directly use the words appearing in the title as keywords, and set the weight as 1; and obtaining the weight of each word by using the word frequency and the word coverage of the words appearing in the text, then arranging the weights in a descending order, and taking at least one keyword with a larger weight value as the content characteristic corresponding to the information B.
Step 302, determining a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content.
In an embodiment of the present invention, after the information recommendation platform obtains the pre-stored content corresponding to the preset source information, it may respectively determine a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content.
It should be noted that, in the embodiment of the present invention, the weight of the first keyword is a weight value corresponding to the same keyword in the information; accordingly, the second keyword weight is a weight value corresponding to the keyword of the pre-stored content. And the first keyword weight and the second keyword weight are weight values corresponding to the same keyword in different information respectively.
Specifically, in the embodiment of the present invention, the first keyword weight and the second keyword weight may be obtained by calculating according to a word frequency factor and a word coverage of the keyword. Further, in an embodiment of the present invention, for the keyword i, the information recommendation platform may obtain the word frequency factor tf of the keyword i by calculating according to the formula (2) and the formula (3) respectivelyiSum word coverage spani。
Figure BDA0001598215710000151
Figure BDA0001598215710000152
Wherein, freiIs the frequency of occurrence of the keyword i in the information, lasiAnd firiRespectively representing the position parameter of the last occurrence and the position parameter of the first occurrence of the keyword in the information, and sum is the sum of all the position parameters of the keyword in the information.
Further, in the embodiment of the present invention, the information recommendation platform calculates the word frequency factor tf of the keyword i according to the formula (2) and the formula (3) respectivelyiSum word coverage spaniThen, according to the word frequency factor and word coverage of the keyword i, the corresponding weight is obtained by calculation of formula (4)i。
weighti＝tfi*spani (4)
And 303, inputting the weight of the first keyword and the weight of the second keyword into a preset content similarity calculation model to obtain the content similarity.
In an embodiment of the invention, after the information recommendation platform determines the first keyword weight corresponding to the information content and the second keyword weight corresponding to the pre-stored content, the information recommendation platform may input the first keyword weight and the second keyword weight into a pre-set content similarity calculation model to obtain the content similarity.
Further, in an embodiment of the present invention, after the information recommendation platform calculates the weight of the first keyword and the weight of the second keyword, that is, after a weight value of the same keyword in one piece of information in the first browsing data and a weight value of the same keyword in another piece of information in the preset source information are respectively calculated and obtained, the content similarity corresponding to two different pieces of information can be obtained by calculating the weight of the first keyword and the weight of the second keyword corresponding to the same keyword.
It should be noted that, in the embodiment of the present invention, the information recommendation platform may calculate the content similarity Con (i, j) between the information i and the information j according to the formula (5) through the keyword k.
Figure BDA0001598215710000161
Wherein Con (i, j) represents the similarity between the information i and the information j, k represents the k-th keyword that is the same as the information i and the information j, L represents the sum of the keywords that are the same as the information i and the information j, wikWeight, w, of the kth keyword in the information i representing successful matching between the information i and the information jjkRepresents the weight of the k-th keyword in the information j that matches the information i and the information j successfully.
Fig. 5 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 5, in an embodiment of the present invention, the method for determining the first recommendation list according to the attribute similarity and the content similarity may include the following steps:
step 401, inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information.
In an embodiment of the present invention, after determining the attribute similarity and the content similarity corresponding to the first browsing data and the preset source information according to the information attribute, the information content, and the preset source information, the information recommendation platform may input the attribute similarity and the content similarity into a preset pre-stored information similarity calculation model, so as to obtain the information similarity between the first browsing data and the preset source information.
Further, in an embodiment of the present invention, after the information recommendation platform respectively calculates and obtains the attribute similarity and the content similarity between the first browsing data and the preset source information, the attribute similarity and the content similarity may be further fused to obtain the information similarity.
It should be noted that, in the embodiment of the present invention, after obtaining the attribute similarity sim (i, j) and the content similarity Con (i, j) between the information i and the information j, the information recommendation platform can calculate and obtain the information similarity sim (i, j) according to the formula (6)final(i，j)。
Figure BDA0001598215710000171
Step 402, determining a first online time corresponding to the predetermined source information.
In an embodiment of the invention, the information recommendation platform may further determine a first online time corresponding to the preset source information after inputting the attribute similarity and the content similarity into the pre-stored information similarity calculation model and calculating to obtain the information similarity between the first browsing data and the preset source information.
It should be noted that, in the embodiment of the present invention, the first online time is a latest occurrence time corresponding to any one of the preset source information. Specifically, in an embodiment of the present invention, the unit of the first online time may be "day".
Step 403, determining a first recommendation list according to the information similarity and the first online time.
In an embodiment of the invention, after the information recommendation platform calculates the information similarity between the first browsing data and the preset source information and the first online time corresponding to the preset source information, the information recommendation platform may determine the first recommendation list according to the information similarity and the first online time.
Further, in an embodiment of the present invention, the information recommendation platform may further adjust the information similarity according to the first online time, then select a plurality of information with highest similarity from the preset source information according to the adjusted information similarity, and create the first recommendation list according to the plurality of information with highest similarity.
Specifically, in the embodiment of the present invention, for the information i and the information j, the information recommendation platform may obtain the adjusted information similarity sim 'according to the formula (7)'final(i，j)。
sim/ final(i,j)＝simfinal(i,j)*e-a (7)
Wherein a is a first online time corresponding to information j similar to information i.
Further, in an embodiment of the present invention, after the information recommendation platform determines, from the preset source information, a plurality of information with higher relative similarity corresponding to each information in the first browsing data, the information recommendation platform may establish the first recommendation list according to the plurality of information, so as to obtain an information recommendation result corresponding to the recommendation object.
The information recommendation method provided by the embodiment of the invention obtains first browsing data corresponding to a recommendation object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, according to the information recommendation method provided by the embodiment of the invention, whether the first browsing data corresponding to the recommended object belongs to the preset source information or not can be determined, then the first recommendation list is obtained according to the information attribute and the information content corresponding to the first browsing data, or the second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommended object, and finally the recommended object is subjected to personalized information recommendation according to the first recommendation list or the second recommendation list, so that the problem that the recommended object without browsing data cannot be subjected to information recommendation can be solved, and further the personalized information recommendation can be performed on any type of recommended object.
Example four
Based on the first embodiment, fig. 6 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 6, in an embodiment of the present invention, when first browsing data does not exist in preset source information, a method for obtaining a second recommendation list according to the first browsing data and attribute information corresponding to a recommendation object may include the following steps:
and 103a, acquiring a target object according to the first browsing data, the attribute information and the pre-stored object identifier.
In an embodiment of the invention, after the information recommendation platform obtains the first browsing data corresponding to the recommended object, if the preset source information does not have the first browsing data, a target object may be obtained according to the first browsing data, the attribute information, and a pre-stored object identifier.
It should be noted that, in an embodiment of the present invention, after the information recommendation platform obtains the first browsing data corresponding to the recommended object, if the preset source information does not have the first browsing data, the information recommendation platform may first obtain attribute information corresponding to the recommended object.
It should be noted that, in the embodiment of the present invention, if the first browsing data does not exist in the preset source information, the first browsing data may be behavior data of the recommended object accessing other website resources. The attribute information may include specific attributes such as a grade, an identity, an occupation, and a value type of the recommended object.
Further, in an embodiment of the present invention, the information recommendation platform may establish an object representation corresponding to the recommended object, that is, the object identification information, according to the first browsing data and the attribute information, so as to determine the target object in the pre-stored object identification according to the identification information of the recommended object.
It should be noted that, in the embodiment of the present invention, the target object is an object corresponding to an object identifier with a high similarity to the object identifier information in the pre-stored object identifiers.
And 103b, acquiring second browsing data corresponding to the target object.
In an embodiment of the invention, after the information recommendation platform obtains the target object according to the first browsing data, the attribute information, and the pre-stored object identifier, the information recommendation platform may obtain second browsing data corresponding to the target object.
It should be noted that, in an embodiment of the present invention, the second browsing data may include at least one piece of information that the target object has browsed, and meanwhile, the second browsing data belongs to the preset source information, that is, the target object has browsed the preset source information.
And 103c, determining a second recommendation list according to the second browsing data.
In an embodiment of the invention, after the information recommendation platform obtains second browsing data corresponding to the target object, the information recommendation platform may determine the second recommendation list according to the second browsing data.
Further, in an embodiment of the present invention, after the information recommendation platform obtains second browsing data corresponding to the target recommendation object, the information recommendation platform may establish the second recommendation list according to at least one piece of information in the second browsing data.
According to the description, through the steps 103a to 103c, the information recommendation platform obtains the target object according to the first browsing data, the attribute information and the pre-stored object identifier; acquiring second browsing data corresponding to the target object; and determining a second recommendation list according to the second browsing data. Therefore, the information recommendation method provided by the embodiment of the invention can determine whether the first browsing data corresponding to the recommended object belongs to the preset source information, then obtain the first recommendation list according to the information attribute and the information content corresponding to the first browsing data, or obtain the second recommendation list according to the first browsing data and the attribute information corresponding to the recommended object, and finally perform personalized information recommendation on the recommended object according to the first recommendation list or the second recommendation list, so that the problem that the recommended object without browsing data cannot perform information recommendation can be solved, the personalized information recommendation on any type of recommended object can be further realized, and the cold start problem during information recommendation can be effectively solved.
EXAMPLE five
Based on the fourth embodiment, fig. 7 is a seventh implementation flow diagram of the information recommendation method according to the third embodiment of the present invention, as shown in fig. 7, in an embodiment of the present invention, the method for acquiring a target object by the information recommendation platform according to the first browsing data, the attribute information, and the pre-stored object identifier may include the following steps:
step 501, obtaining identification information corresponding to the recommended object according to the first browsing data and the attribute information.
In an embodiment of the invention, after the information recommendation platform obtains the attribute information corresponding to the object, the information recommendation platform may obtain the identification information corresponding to the recommended object according to the first browsing data and the attribute information.
It should be noted that, in an embodiment of the present invention, the identification information is an object representation created by the information recommendation platform according to the first browsing data of the recommended object and the attribute information.
Specifically, in an embodiment of the present invention, the information recommendation platform may obtain an index parameter of the recommended object according to the first browsing data and the attribute information, and further perform a dummy transformation on the index parameter, so as to obtain the identification information corresponding to the recommended object.
Step 502, inputting the identification information and the pre-stored object identification into a preset object similarity calculation model to obtain the object similarity.
In an embodiment of the present invention, after obtaining the identification information corresponding to the recommended object according to the historical behavior record and the attribute information, the information recommendation platform may input the identification information and the pre-stored object identifier into a preset object similarity calculation model, and calculate to obtain an object similarity.
It should be noted that, in the embodiment of the present invention, the preset object similarity calculation model is used for calculating and obtaining the similarity between different objects according to the identification information corresponding to the different objects.
Further, in the embodiment of the present invention, for the object i and the object j, the information recommendation platform may calculate and obtain the object similarity sim (i, j) between the object i and the object j according to the formula (8).
Figure BDA0001598215710000211
Wherein, XiAnd XjRespectively representing the identification information of object i and object j.
And step 503, determining the target object according to the object similarity.
In an embodiment of the present invention, after the information recommendation platform inputs the identification information and the pre-stored object identification into a preset object similarity calculation model, and calculates and obtains the object similarity, the information recommendation platform may determine the target object that is similar to the recommended object from the pre-stored object according to the object similarity.
Fig. 8 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 8, in an embodiment of the present invention, the method for determining the first recommendation list by the information recommendation platform according to the second browsing data may include the following steps:
step 601, determining a second online time corresponding to the second browsing data.
In an embodiment of the invention, after the information recommendation platform obtains the second browsing data corresponding to the target object, the information recommendation platform can determine a second online time corresponding to the second browsing data.
In an embodiment of the invention, the second online time is a latest time of occurrence corresponding to any information in the second browsing data. Specifically, in an embodiment of the present invention, the unit of the second online time may be "day".
Step 602, determining a second recommendation list according to the second browsing data and the second online time.
In an embodiment of the invention, after determining the second online time corresponding to the second browsing data, the information recommendation platform may further determine the second recommendation list according to the second browsing data and the second online time.
Further, in an embodiment of the present invention, the information recommendation platform may further select a plurality of pieces of information with higher interest degrees from the second browsing data according to the second online time, and create the second recommendation list according to the plurality of pieces of information with higher interest degrees.
Specifically, in the embodiment of the present invention, the information recommendation platform may obtain the interest level exc of the object i in the information j according to the formula (9)ij。
excij＝sim(i,j)*e-a (9)
Wherein a is the second on-line time corresponding to the information j.
Further, in an embodiment of the present invention, after the information recommendation platform determines a plurality of pieces of information with higher interest of the recommendation object from the second browsing data, the information recommendation platform may establish the second recommendation list according to the plurality of pieces of information, so as to obtain an information recommendation result corresponding to the recommendation object.
The information recommendation method provided by the embodiment of the invention obtains first browsing data corresponding to a recommendation object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, the information recommendation method provided by the embodiment of the invention can determine whether the first browsing data corresponding to the recommended object belongs to the preset source information, then obtain the first recommendation list according to the information attribute and the information content corresponding to the first browsing data, or obtain the second recommendation list according to the first browsing data and the attribute information corresponding to the recommended object, and finally perform personalized information recommendation on the recommended object according to the first recommendation list or the second recommendation list, so that the problem that the recommended object without browsing data cannot perform information recommendation can be solved, the personalized information recommendation on any type of recommended object can be further realized, and the cold start problem during information recommendation can be effectively solved.
EXAMPLE six
Based on the third embodiment, fig. 9 is a ninth implementation flow diagram of an information recommendation method according to an embodiment of the present invention, as shown in fig. 9, in an embodiment of the present invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, that is, after step 101, the method for the information recommendation platform to perform information recommendation may further include the following steps:
105, acquiring a third recommendation list according to a preset heat type; wherein, the preset heat type is used for recommending the information according to the heat of the information.
In an embodiment of the present invention, after the information recommendation platform obtains the first browsing data corresponding to the recommendation object, the information recommendation platform may obtain a third recommendation list according to a preset popularity type.
It should be noted that, in the embodiment of the present invention, the preset popularity type may be an information recommendation method that the information recommendation platform recommends according to popularity of information.
Further, in an embodiment of the invention, the third recommendation list may be a recommendation list including a plurality of information obtained after the information recommendation platform performs information recommendation according to the preset popularity recommendation type.
Fig. 10 is a schematic flow chart illustrating an implementation process of an information recommendation method according to an embodiment of the present invention, as shown in fig. 10, in an embodiment of the present invention, a method for the information recommendation platform to obtain the third recommendation list according to the preset heat type may include the following steps:
step 105a, obtaining the access amount corresponding to the preset source information.
In an embodiment of the invention, the information recommendation platform may first obtain the access amount corresponding to the predetermined source information.
It should be noted that, in the embodiment of the present invention, the access amount corresponding to the preset source information may be at least one historical access data corresponding to at least one information in the preset source information.
And 105b, inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to the preset source information.
In an embodiment of the present invention, after obtaining the access amount corresponding to the preset source information, the information recommendation platform may input the access amount and the first online time into a preset heat calculation model, so as to obtain a heat parameter corresponding to the preset source information.
Further, in an embodiment of the present invention, the predetermined heat calculation model is used for obtaining the heat parameter of the information according to the access amount and the online time. Specifically, in the embodiment of the present invention, the information recommendation platform may obtain the hotness parameter HOT according to the formula (10).
HOT＝READ_COUNT*e-a (10)
Wherein, READ _ COUNT is the access amount corresponding to the information,ais the first online time corresponding to the information.
And 105c, acquiring a third recommendation list according to the heat parameter.
In an embodiment of the present invention, the information recommendation platform may obtain a third recommendation list according to the heat parameter after inputting the access amount and the first online time into a preset heat calculation model to obtain the heat parameter corresponding to the preset source information.
It should be noted that, in the embodiment of the present invention, the third recommendation list is determined according to the access heat of the information, and the information in the third recommendation list is a plurality of information with higher heat parameters and more online time.
According to the description, through the steps 105a to 105c, the information recommendation platform obtains the access amount corresponding to the preset source information; inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to preset source information; and acquiring a third recommendation list according to the heat parameter. Therefore, the information recommendation method provided in the embodiment of the present invention may determine whether the first browsing data corresponding to the recommended object belongs to the preset source information, then obtain the first recommendation list according to the information attribute and the information content corresponding to the first browsing data, or obtain the second recommendation list according to the first browsing data and the attribute information corresponding to the recommended object, and finally perform personalized information recommendation on the recommended object according to the first recommendation list or the second recommendation list, so as to solve the problem that the recommended object without browsing data cannot perform information recommendation, and further perform personalized information recommendation on any type of recommended object, and further, the present invention may combine the third recommendation list obtained according to the preset popularity type with the first recommendation list to determine the information recommendation result, therefore, the problem of sparse information can be solved, and the richness of information recommendation is greatly improved.
In an embodiment of the present invention, further, the method for the information recommendation platform to recommend information to the recommendation object according to the first recommendation list and the second recommendation list may specifically be to recommend information to the recommendation object according to the first recommendation list and the third recommendation list; or recommending information to the recommendation object according to the second recommendation list and the third recommendation list.
In an embodiment of the present invention, further, fig. 11 is a schematic view illustrating an implementation flow of an information recommendation method according to an embodiment of the present invention, as shown in fig. 11, in an embodiment of the present invention, the method for the information recommendation platform to recommend information to a recommendation object according to the first recommendation list and the third recommendation list may include the following steps:
step 701, determining a first information quantity corresponding to the first recommendation list.
In an embodiment of the invention, the information recommendation platform may determine the first information amount corresponding to the first recommendation list first.
It should be noted that, in the embodiment of the present invention, for a website with a small information updating amount and a slow information updating speed, the amount of information in the first recommendation list may be less.
Step 702, if the first information quantity is smaller than the preset quantity threshold, the third recommendation list is supplemented to the first recommendation list to recommend information to the recommendation object.
In an embodiment of the invention, after determining the first information quantity corresponding to the first recommendation list, if the first information quantity is smaller than a preset quantity threshold, the information recommendation platform may supplement the third recommendation list to the first recommendation list, so as to determine the information recommendation result.
Further, in an embodiment of the present invention, the information recommendation platform may compare the first information quantity with the preset quantity threshold, and if the first information quantity is smaller than the preset quantity threshold, that is, the first information quantity in the first recommendation list is insufficient, the information recommendation platform may supplement the information in the third recommendation list to the first recommendation list, so as to obtain the information recommendation result.
And 703, directly recommending information to the recommendation object according to the first recommendation list if the first information quantity is greater than or equal to the preset quantity threshold.
In an embodiment of the invention, after the information recommendation platform determines the first information quantity corresponding to the first recommendation list, if the first information quantity is greater than or equal to a preset quantity threshold, the information recommendation platform may directly determine the first recommendation list as the information recommendation result.
Further, in an embodiment of the present invention, the information recommendation platform may compare the first information quantity with the preset quantity threshold, and if the first information quantity is greater than or equal to the preset quantity threshold, that is, the first information quantity in the first recommendation list is sufficient, the information recommendation platform may directly determine the first recommendation list as the information recommendation result without supplementing the information in the third recommendation list to the first recommendation list.
In an embodiment of the present invention, further, fig. 12 is a flowchart illustrating an implementation flow of a method for recommending information according to an embodiment of the present invention as a twelfth, as shown in fig. 12, in an embodiment of the present invention, the method for recommending information to a recommended object by the information recommendation platform according to the second recommendation list and the third recommendation list may include the following steps:
step 801, determining a second information quantity corresponding to the second recommendation list.
In an embodiment of the invention, the information recommendation platform may determine the second information amount corresponding to the second recommendation list first.
It should be noted that, in the embodiment of the present invention, for the website with small information updating amount and slow information updating speed, the amount of information in the second recommendation list may be less.
Step 802, if the second information quantity is smaller than the preset quantity threshold, the third recommendation list is supplemented to the second recommendation list to recommend information to the recommendation object.
In an embodiment of the invention, after determining the second information quantity corresponding to the second recommendation list, if the second information quantity is smaller than a preset quantity threshold, the information recommendation platform may supplement the third recommendation list to the second recommendation list, so as to determine the information recommendation result.
Further, in an embodiment of the present invention, the information recommendation platform may compare the second information quantity with the preset quantity threshold, and if the second information quantity is smaller than the preset quantity threshold, that is, the second information quantity in the second recommendation list is insufficient, the information recommendation platform may supplement the information in the third recommendation list to the second recommendation list, so as to obtain the information recommendation result.
And 803, if the second information quantity is greater than or equal to the preset quantity threshold, directly recommending information to the recommendation object according to the second recommendation list.
In an embodiment of the invention, after the information recommendation platform determines the second information quantity corresponding to the second recommendation list, if the second information quantity is greater than or equal to a preset quantity threshold, the information recommendation platform may directly determine the second recommendation list as the information recommendation result.
Further, in an embodiment of the present invention, the information recommendation platform may compare the second information quantity with the preset quantity threshold, and if the second information quantity is greater than or equal to the preset quantity threshold, that is, the second information quantity in the second recommendation list is sufficient, the information recommendation platform may directly determine the second recommendation list as the information recommendation result without supplementing the information in the third recommendation list to the second recommendation list.
The information recommendation method provided by the embodiment of the invention obtains first browsing data corresponding to a recommendation object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, the information recommendation method provided in the embodiment of the present invention may determine whether the first browsing data corresponding to the recommended object belongs to the preset source information, then obtain the first recommendation list according to the information attribute and the information content corresponding to the first browsing data, or obtain the second recommendation list according to the first browsing data and the attribute information corresponding to the recommended object, and finally perform personalized information recommendation on the recommended object according to the first recommendation list or the second recommendation list, so as to solve the problem that the recommended object without browsing data cannot perform information recommendation, and further perform personalized information recommendation on any type of recommended object, and further, the present invention may combine the third recommendation list obtained according to the preset popularity type with the first recommendation list to determine the information recommendation result, therefore, the problem of sparse information can be solved, and the richness of information recommendation is greatly improved.
EXAMPLE seven
Based on the same inventive concept of the first to sixth embodiments, fig. 13 is a schematic structural diagram of a composition of an information recommendation platform according to the first embodiment of the present invention, as shown in fig. 13, in an embodiment of the present invention, an information recommendation platform 1 includes: an acquisition unit 11 and a recommendation unit 12.
The acquiring unit 11 is configured to acquire first browsing data corresponding to a recommended object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; and when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and the attribute information corresponding to the recommendation object.
The determining unit 12 is configured to recommend information to the recommendation object according to the first recommendation list or the second recommendation list.
In the embodiment of the present invention, further, the obtaining unit 11 is further configured to obtain a third recommendation list according to a preset heat type after obtaining the first browsing data corresponding to the recommendation object; wherein, the preset heat type is used for recommending the information according to the heat of the information.
In the embodiment of the present invention, further, the recommending unit 12 is further configured to recommend information to the recommended object according to the first recommendation list and the third recommendation list; or recommending information to the recommendation object according to the second recommendation list and the third recommendation list.
The obtaining unit 11 and the recommending unit 12 provided by the embodiment of the present invention can be implemented in the form of program codes by executing corresponding functions by a processor in a server; of course, the implementation can also be realized through a specific logic circuit; in the process of the embodiment, the Processor may be a Central Processing Unit (CPU), a Microprocessor Unit (MPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), or the like, and the information recommendation platform further includes a memory, which may be a storage device having a physical form, such as a memory bank or a TF card, or a circuit having a storage function, such as a Random Access Memory (RAM), a FIFO memory, or the like.
Based on the same inventive concept of the first to sixth embodiments, fig. 14 is a schematic structural diagram of a second composition of the information recommendation platform according to the second embodiment of the present invention, and as shown in fig. 14, the information recommendation platform 1 according to the second embodiment of the present invention may include a processor 13, a memory 14 storing executable instructions of the processor 13, and a communication bus 15 for connecting the processor 13 and the memory 14.
In the process of the Specific embodiment, the PRocessoR 13 may be at least one of an Application Specific IntegRated CiRcuit (ASIC), a Digital Signal PRocessoR (DSP), a Digital Signal PRocessing Device (DSPD), a PRogRammable Logic Device (PLD), a Field PRogRammable Gate ARRay (FPGA), a CentRal PRocessing Unit (CPU), a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronic devices used to implement the processor functions described above may be other devices, and embodiments of the present invention are not limited in particular.
In the embodiment of the present invention, the communication bus 15 is used for realizing connection communication between the processor 13 and the memory 14; the processor 13 is configured to execute the information recommendation program stored in the memory 14 to implement the following steps:
acquiring first browsing data corresponding to a recommended object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
The information recommendation platform provided by the embodiment of the invention acquires first browsing data corresponding to a recommended object; when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data; when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object; and recommending information to the recommendation object according to the first recommendation list or the second recommendation list. Therefore, according to the information recommendation method, the information recommendation platform and the computer-readable storage medium provided by the embodiment of the invention, whether the first browsing data corresponding to the recommended object belongs to the preset source information or not can be determined, then the first recommendation list is obtained according to the information attribute and the information content corresponding to the first browsing data, or the second recommendation list is obtained according to the first browsing data and the attribute information corresponding to the recommended object, and finally the personalized information recommendation is performed on the recommended object according to the first recommendation list or the second recommendation list, so that the problem that the information recommendation cannot be performed on the recommended object without the browsing data can be solved, and the personalized information recommendation can be performed on any type of recommended object. Moreover, the method is simple and convenient to realize, convenient to popularize and wide in application range.
The embodiment of the present invention provides a computer-readable storage medium, where one or more programs are stored, where the one or more programs are executable by one or more processors and applied to an information recommendation platform, and when the programs are executed by the processors, the method according to the first to sixth embodiments is implemented.
Specifically, the program instructions corresponding to an information recommendation method in the present embodiment may be stored on a storage medium such as an optical disc, a hard disc, a usb disk, etc., and when the program instructions corresponding to an information recommendation method in the storage medium are read or executed by an electronic device, the method includes the following steps:
acquiring first browsing data corresponding to a recommended object;
when the preset source information has first browsing data, acquiring a first recommendation list according to information attributes and information contents corresponding to the first browsing data;
when the preset source information does not have the first browsing data, acquiring a second recommendation list according to the first browsing data and attribute information corresponding to the recommendation object;
and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims (15)

1. An information recommendation method, the method comprising:
acquiring first browsing data corresponding to a recommended object;
determining attribute similarity corresponding to the first browsing data and preset source information according to information attributes and the preset source information; determining content similarity corresponding to the first browsing data and the preset source information according to information content and the preset source information; the content similarity is obtained according to the weight of the first keyword and the weight of the second keyword; the first keyword weight and the second keyword weight are obtained through calculation according to the word frequency factor and the word coverage;
inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information;
determining a first online time corresponding to the preset source information;
adjusting the information similarity according to the first online time, wherein the obtained adjusted information similarity is as follows:
sim/ final(i,j)＝simfinal(i,j)*e-α (1)，
wherein, simfinal(i, j) is the information similarity; alpha is a first online time corresponding to information j similar to information i; sim/ final(i, j) is the adjusted information similarity;
determining a first recommendation list according to the adjusted information similarity;
acquiring a target object according to the first browsing data, the attribute information and the pre-stored object identifier; wherein the attribute information comprises the age, identity, occupation and value type of the recommended object;
acquiring second browsing data corresponding to the target object;
determining a second recommendation list according to the second browsing data;
and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
2. The method of claim 1, wherein the determining the attribute similarity corresponding to the first browsing data and the predetermined source information according to the information attribute and the predetermined source information comprises:
acquiring a pre-stored attribute corresponding to the preset source information;
and inputting the information attributes and the pre-stored attributes into a preset attribute similarity calculation model to obtain the attribute similarity.
3. The method of claim 1, wherein the determining the content similarity corresponding to the first browsing data and the predetermined source information according to the information content and the predetermined source information comprises:
acquiring prestored contents corresponding to the preset source information;
determining a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content;
and inputting the first keyword weight and the second keyword weight into a preset content similarity calculation model to obtain the content similarity.
4. The method according to claim 1, wherein the obtaining a target object according to the first browsing data, the attribute information, and a pre-stored object identifier comprises:
acquiring identification information corresponding to the recommended object according to the first browsing data and the attribute information;
inputting the identification information and the pre-stored object identification into a preset object similarity calculation model to obtain object similarity;
and determining the target object according to the object similarity.
5. The method of claim 4, wherein determining the second recommendation list based on the second browsing data comprises:
determining second online time corresponding to the second browsing data;
and determining the second recommendation list according to the second browsing data and the second online time.
6. The method according to claim 1, wherein after the first browsing data corresponding to the recommended object is obtained, the method further comprises:
acquiring a third recommendation list according to a preset heat type; the preset popularity type is used for recommending the information according to the popularity of the information.
7. The method of claim 6, wherein the obtaining the third recommendation list according to the preset heat type comprises:
acquiring the access amount corresponding to the preset source information;
inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to the preset source information;
and acquiring the third recommendation list according to the heat parameter.
8. The method of claim 7, wherein the recommending information to the recommendation object according to the first recommendation list or the second recommendation list comprises:
according to the first recommendation list and the third recommendation list, recommending information to the recommendation object; alternatively, the first and second electrodes may be,
and recommending information to the recommended object according to the second recommendation list and the third recommendation list.
9. The method of claim 8, wherein the recommending information to the recommendation object according to the first recommendation list and the third recommendation list comprises:
determining a first information quantity corresponding to the first recommendation list;
if the first information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the first recommendation list so as to recommend information to the recommendation object;
and if the first information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the first recommendation list.
10. The method of claim 8, wherein the recommending information to the recommendation object according to the second recommendation list and the third recommendation list comprises:
determining a second information quantity corresponding to the second recommendation list;
if the second information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the second recommendation list so as to recommend information to the recommendation object;
and if the second information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the second recommendation list.
11. An information recommendation platform, comprising: an acquisition unit and a recommendation unit,
the acquisition unit is used for acquiring first browsing data corresponding to the recommended object; determining attribute similarity corresponding to the first browsing data and preset source information according to information attributes and the preset source information; determining content similarity corresponding to the first browsing data and the preset source information according to information content and the preset source information; the content similarity is obtained according to the weight of the first keyword and the weight of the second keyword; the first keyword weight and the second keyword weight are obtained through calculation according to the word frequency factor and the word coverage; inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information; determining a first online time corresponding to the preset source information; adjusting the information similarity according to the first online time, wherein the obtained adjusted information similarity is as follows:
sim/ final(i,j)＝simfinal(i,j)*e-α (1)，
wherein, simfinal(i, j) is the information similarity; alpha is a first online time corresponding to information j similar to information i; sim/ final(i, j) is the adjusted information similarity;
determining a first recommendation list according to the adjusted information similarity; acquiring a target object according to the first browsing data, the attribute information and the pre-stored object identifier; wherein the attribute information comprises the age, identity, occupation and value type of the recommended object; acquiring second browsing data corresponding to the target object; determining a second recommendation list according to the second browsing data;
and the recommending unit is used for recommending information to the recommending object according to the first recommending list or the second recommending list.
12. The information recommendation platform of claim 11,
the acquisition unit is further configured to acquire a third recommendation list according to a preset popularity type after acquiring first browsing data corresponding to the recommendation object; the preset heat type is used for recommending information according to the heat of the information;
the recommending unit is further used for recommending information to the recommending object according to the first recommending list and the third recommending list; or recommending information to the recommendation object according to the second recommendation list and the third recommendation list.
13. An information recommendation platform, comprising: a processor, a memory, and a communication bus; the communication bus is used for realizing connection communication between the processor and the memory; the processor is used for executing the data transmission program stored in the memory so as to realize the following steps:
acquiring first browsing data corresponding to a recommended object;
determining attribute similarity corresponding to the first browsing data and preset source information according to information attributes and the preset source information; determining content similarity corresponding to the first browsing data and the preset source information according to information content and the preset source information; determining a first recommendation list according to the attribute similarity and the content similarity; and acquiring the pre-stored attribute corresponding to the preset source information; inputting the information attribute and the pre-stored attribute into a preset attribute similarity calculation model to obtain the attribute similarity; and acquiring prestored contents corresponding to the preset source information; determining a first keyword weight corresponding to the information content and a second keyword weight corresponding to the pre-stored content; the first keyword weight and the second keyword weight are obtained through calculation according to a word frequency factor and word coverage; inputting the first keyword weight and the second keyword weight into a preset content similarity calculation model to obtain the content similarity; inputting the attribute similarity and the content similarity into a pre-stored information similarity calculation model to obtain the information similarity between the first browsing data and the preset source information; determining a first online time corresponding to the preset source information; and adjusting the information similarity according to the first online time, wherein the obtained adjusted information similarity is as follows:
sim/ final(i,j)＝simfinal(i,j)*e-α (1)，
wherein, simfinal(i, j) is the information similarity; alpha is a first online time corresponding to information j similar to information i; sim/ final(i, j) is the adjusted information similarity;
determining the first recommendation list according to the adjusted information similarity;
acquiring a target object according to the first browsing data, the attribute information and the pre-stored object identifier; acquiring second browsing data corresponding to the target object; determining a second recommendation list according to the second browsing data; acquiring identification information corresponding to the recommended object according to the first browsing data and the attribute information; wherein the attribute information comprises the age, identity, occupation and value type of the recommended object; inputting the identification information and the pre-stored object identification into a preset object similarity calculation model to obtain object similarity; and determining the target object according to the object similarity; determining a second online time corresponding to the second browsing data; determining the second recommendation list according to the second browsing data and the second online time;
and recommending information to the recommendation object according to the first recommendation list or the second recommendation list.
14. The information recommendation platform of claim 13,
the processor is further configured to obtain a third recommendation list according to a preset popularity type after obtaining first browsing data corresponding to the recommended object; the preset heat type is used for recommending information according to the heat of the information;
the processor is further specifically configured to obtain an access amount corresponding to the preset source information; inputting the access amount and the first online time into a preset heat calculation model to obtain a heat parameter corresponding to the preset source information; acquiring the third recommendation list according to the heat parameter; and recommending information to the recommended object according to the first recommendation list and the third recommendation list; or, according to the second recommendation list and the third recommendation list, recommending information to the recommendation object; determining a first information quantity corresponding to the first recommendation list; if the first information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the first recommendation list so as to recommend information to the recommendation object; if the first information quantity is larger than or equal to the preset quantity threshold value, information recommendation is directly carried out on the recommendation object according to the first recommendation list; determining a second information quantity corresponding to the second recommendation list; if the second information quantity is smaller than a preset quantity threshold value, supplementing the third recommendation list to the second recommendation list so as to recommend information to the recommendation object; and if the second information quantity is larger than or equal to the preset quantity threshold value, directly recommending information to the recommendation object according to the second recommendation list.
15. A computer readable storage medium, characterized in that the computer readable storage medium stores one or more programs which are executable by one or more processors to implement the method of any one of claims 1 to 10.
CN201810214568.1A 2018-03-15 2018-03-15 Information recommendation method, platform and computer readable storage medium Active CN110309410B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810214568.1A CN110309410B (en) 2018-03-15 2018-03-15 Information recommendation method, platform and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810214568.1A CN110309410B (en) 2018-03-15 2018-03-15 Information recommendation method, platform and computer readable storage medium

Publications (2)

Publication Number Publication Date
CN110309410A CN110309410A (en) 2019-10-08
CN110309410B true CN110309410B (en) 2022-04-29

Family

ID=68073838

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810214568.1A Active CN110309410B (en) 2018-03-15 2018-03-15 Information recommendation method, platform and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN110309410B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111368202B (en) * 2020-03-06 2023-09-19 咪咕文化科技有限公司 Search recommendation method and device, electronic equipment and storage medium
CN111429232A (en) * 2020-04-12 2020-07-17 中信银行股份有限公司 Product recommendation method and device, electronic equipment and computer-readable storage medium
CN111737558A (en) * 2020-05-21 2020-10-02 苏宁金融科技(南京)有限公司 Information recommendation method and device and computer readable storage medium
CN113836404A (en) * 2021-09-09 2021-12-24 武汉卓尔数字传媒科技有限公司 Object recommendation method and device, electronic equipment and computer-readable storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559315A (en) * 2013-11-20 2014-02-05 上海华勤通讯技术有限公司 Information screening and pushing method and device
CN103678647A (en) * 2013-12-20 2014-03-26 Tcl集团股份有限公司 Method and system for recommending information
CN104809154A (en) * 2015-03-19 2015-07-29 百度在线网络技术（北京）有限公司 Method and device for recommending information
CN107657004A (en) * 2017-09-21 2018-02-02 广州华多网络科技有限公司 Video recommendation method, system and equipment

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140331142A1 (en) * 2013-05-06 2014-11-06 Tencent Technology (Shenzhen) Company Limited Method and system for recommending contents
CN105913273A (en) * 2015-12-14 2016-08-31 乐视网信息技术（北京）股份有限公司 Intelligent recommendation all-terminal display method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559315A (en) * 2013-11-20 2014-02-05 上海华勤通讯技术有限公司 Information screening and pushing method and device
CN103678647A (en) * 2013-12-20 2014-03-26 Tcl集团股份有限公司 Method and system for recommending information
CN104809154A (en) * 2015-03-19 2015-07-29 百度在线网络技术（北京）有限公司 Method and device for recommending information
CN107657004A (en) * 2017-09-21 2018-02-02 广州华多网络科技有限公司 Video recommendation method, system and equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于内容的相关书籍推荐技术研究;商雪晶;《中国优秀硕士学位论文全文数据库》;20110515(第5期);第18-19页第3.2部分段、第31页第4.2部分-第35页第4.3部分以及第41页第5.3.1部分 *

Also Published As

Publication number Publication date
CN110309410A (en) 2019-10-08

Similar Documents

Publication Publication Date Title
US10860811B2 (en) Method and device for generating review article of hot news, and terminal device
CN109885770B (en) Information recommendation method and device, electronic equipment and storage medium
CN110309410B (en) Information recommendation method, platform and computer readable storage medium
CN105808685B (en) Promotion information pushing method and device
WO2017181612A1 (en) Personalized video recommendation method and device
TWI542391B (en) Recommendation method, device, and server
WO2020048084A1 (en) Resource recommendation method and apparatus, computer device, and computer-readable storage medium
CN108509479B (en) Entity recommendation method and device, terminal and readable storage medium
US20140172642A1 (en) Analyzing commodity evaluations
CN104636371A (en) Information recommendation method and device
CN110334356B (en) Article quality determining method, article screening method and corresponding device
WO2012060866A1 (en) Determination of category information using multiple stages
US20180285331A1 (en) Method, server, browser, and system for recommending text information
CN111639255B (en) Recommendation method and device for search keywords, storage medium and electronic equipment
EP2836974A2 (en) Method, web server and web browser of providing information
CN109241451B (en) Content combination recommendation method and device and readable storage medium
EP3818492B1 (en) Communication via simulated user
CN111506831A (en) Collaborative filtering recommendation module and method, electronic device and storage medium
CN111522886B (en) Information recommendation method, terminal and storage medium
CN108536786A (en) A kind of information recommendation method, device, server and storage medium
CN111552884A (en) Method and apparatus for content recommendation
CN107368489A (en) A kind of information data processing method and device
CN104102662A (en) Method and device for determining interest and preference similarity of users
CN114581165A (en) Product recommendation method, device, computer storage medium and system
CN111723280B (en) Information processing method and device, storage medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220701

Address after: 610041 China (Sichuan) pilot Free Trade Zone, Chengdu, Sichuan

Patentee after: China Mobile (Chengdu) information and Communication Technology Co.,Ltd.

Patentee after: China Mobile Communications Group Co., Ltd

Address before: 100032 No. 29, Finance Street, Beijing, Xicheng District

Patentee before: CHINA MOBILE COMMUNICATIONS GROUP Co.,Ltd.

TR01 Transfer of patent right