CN111523049B - Method, device, storage medium and processor for determining authority value of object - Google Patents
Method, device, storage medium and processor for determining authority value of object Download PDFInfo
- Publication number
- CN111523049B CN111523049B CN202010296679.9A CN202010296679A CN111523049B CN 111523049 B CN111523049 B CN 111523049B CN 202010296679 A CN202010296679 A CN 202010296679A CN 111523049 B CN111523049 B CN 111523049B
- Authority
- CN
- China
- Prior art keywords
- determining
- authority
- interaction
- text
- value
- 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
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
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The invention discloses a method, a device, a storage medium and a processor for determining an authority value of an object. Wherein the method comprises the following steps: acquiring a first object set and a second object set, wherein the authority value of each first object in the first object set reaches a target value; acquiring at least one text published by each second object in the second object set; the method comprises the steps of obtaining an interaction result obtained by each first object interacting with each text; and determining the authority value of each second object based on the obtained interaction result. The invention solves the technical problem that the authority value of the object is not effectively determined.
Description
Technical Field
The present invention relates to the field of computers, and in particular, to a method, an apparatus, a storage medium, and a processor for determining an authority value of an object.
Background
Currently, with the development of the internet, various web communities having a common preference or a common goal have arisen, and most of the web communities now appear in the form of Application (APP) of a terminal. People can learn and communicate in the community. Typically, web communities are distinguished to include partial learning communities in which most people learn and acquire knowledge. For this community, most recommendation algorithms only take into account the commercialization of user liveness, etc., and do not determine the authority value of the user object of the published text (e.g., article) to consider as a feature of the recommendation algorithm.
However, if the authority of the user object is not considered in the recommendation algorithm, but only the popularity and click rate of the content are considered, then some articles which are meaningless from the knowledge perspective or even not in good condition are given higher recommendation weights. Thus, although high liveness may be brought about in a short period of time for a community, it is detrimental to the authority of the community in a long term.
Aiming at the technical problem that the authority value of the object is not effectively determined in the prior art, no effective solution is proposed at present.
Disclosure of Invention
The embodiment of the invention provides a method, a device, a storage medium and a processor for determining an authority value of an object, which are used for at least solving the technical problem that the authority value of the object is not effectively determined.
According to one aspect of the embodiment of the invention, a method for determining an authority value of an object is provided. The method may include: acquiring a first object set and a second object set, wherein the authority value of each first object in the first object set reaches a target value; acquiring at least one text published by each second object in the second object set; the method comprises the steps of obtaining an interaction result obtained by each first object interacting with each text; and determining the authority value of each second object based on the obtained interaction result.
Optionally, determining an authority value of each second object based on the obtained interaction result includes: determining authority values of each text based on the obtained interaction result; an authority value for each second object is determined based on the authority value for each text.
Optionally, determining an authority value of each text based on the obtained interaction result includes: acquiring a first number of interaction results in the obtained interaction results, wherein each interaction result in the first number of interaction results is larger than any interaction result except the first number of interaction results in the obtained interaction results; an authoritative value for each text is determined based on the first number of interaction results.
Optionally, determining an authoritative value for each text based on the first number of interaction results includes: and determining an average value of the first number of interaction results as an authority value of each text.
Optionally, determining the authority value of each second object based on the authority value of each text includes: determining a second number of authority values from the authority values of the at least one text, wherein each of the second number of authority values is greater than any one of the authority values of the at least one text other than the second number of authority values; an authority value for each second object is determined based on the second number of authority values.
Optionally, determining an authority value for each second object based on the second number of authority values includes: and determining the average value of the authority values of the second number as the authority value of each second object.
Optionally, obtaining an interaction result obtained by each first object interacting with each text includes: acquiring the field correlation degree of each first object and each text; the interaction strength of each first object on each text is obtained; and determining an interaction result based on the field relevance and the interaction strength.
Optionally, acquiring the domain relevance of each first object and each text includes: determining a topic vector for each text; obtaining the similarity between the topic vector of each text and the topic vector of at least one text published by each first object respectively to obtain at least one similarity; the maximum similarity of the at least one similarity is determined as the domain correlation.
Optionally, acquiring the interaction strength of each first object on each text includes: determining the historical interaction behavior of each first object on each text; and determining the interaction strength according to the weight corresponding to the historical interaction behavior.
Optionally, determining the interaction result based on the domain relevance and the interaction strength includes: and determining the product of the field correlation and the interaction strength as an interaction result.
Optionally, after determining the authority value of each second object based on the obtained interaction result, the method further comprises: and pushing the text published by each second object to the target community based on the authority value of each second object.
According to another aspect of the embodiment of the invention, a device for determining the authority value of the object is also provided. The apparatus may include: the first acquisition unit is used for acquiring a first object set and a second object set, wherein the authority value of each first object in the first object set reaches a target value; the second acquisition unit is used for acquiring at least one text published by each second object in the second object set; the third acquisition unit is used for acquiring an interaction result obtained by each first object interacting with each text; and the determining unit is used for determining the authority value of each second object based on the obtained interaction result.
According to another aspect of the embodiments of the present invention, there is also provided a storage medium. The storage medium comprises a stored program, wherein the device in which the storage medium is controlled to execute the method for determining the authority value of the object according to the embodiment of the invention when the program runs.
According to another aspect of an embodiment of the present invention, there is also provided a processor. The processor is configured to run a program, where the program executes a method for determining an authority value of an object according to an embodiment of the present invention.
In the embodiment of the invention, a first object set and a second object set are acquired, wherein the authority value of each first object in the first object set reaches a target value; acquiring at least one text published by each second object in the second object set; the method comprises the steps of obtaining an interaction result obtained by each first object interacting with each text; the authority value of each second object is determined based on the obtained interaction result, that is, the authority value of the second object is calculated according to the interaction result of each first object and the second object, even if the second object cannot pass through authentication, the authority value is distinguished in authority, so that the technical problem that the authority value of the object is not effectively determined is solved, the technical effect of effectively determining the authority value of the object is achieved, and a recommendation algorithm of texts is better acted.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of a method of determining authority values of an object according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method of determining authority values of another object according to an embodiment of the present invention; and
FIG. 3 is a schematic diagram of an apparatus for determining authority values of an object according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
In accordance with an embodiment of the present invention, there is provided an embodiment of a method of determining authority values of an object, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and that although a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
FIG. 1 is a flow chart of a method of determining authority values of an object according to an embodiment of the present invention. As shown in fig. 1, the method may include the steps of:
step S102, a first object set and a second object set are obtained, wherein the authority value of each first object in the first object set reaches a target value.
In the technical solution provided in the above step S102 of the present invention, a first object set is obtained, where the first object set may include at least one first object, and an authority value of each first object reaches a target value, for example, an authority value a ranges from 0< = a < = 1, where the target value may be 1, where the first object set may be represented by a set Q, each first object may also be called an authority user, and from the first object set, the authority value may be called an authority user set, where the authority value is used to represent authority of a corresponding first object in a target community, and may also be called authority degree.
Alternatively, the embodiment may define the at least one first object, i.e. the authenticated user, in a qualification audit manner and use it as a seed user. It should be noted that the qualification audit tends to be strict, and authority of the user passing the qualification audit needs to be guaranteed, and it is preferable that the knowledge of the at least one first object that is defined can cover all knowledge directions of the target community where the first object is located.
The second object set is obtained, and the second object set can be represented by a set N, wherein each second object is a user object of which the authority value is to be determined, can be other user objects of a target community except the first object set, and can be called a non-authentication user or a non-seed user.
Alternatively, the first set of objects of this embodiment belongs to the same domain as the second set of objects, the first set of objects may comprise at least one expert in the domain, and the second set of objects may comprise at least one ordinary user of the domain.
Step S104, at least one text published by each second object in the second object set is acquired.
In the technical solution provided in the above step S104 of the present invention, after the first object set and the second object set are acquired, at least one text published by each second object in the second object set may be acquired, where the at least one text may include, but is not limited to, an article and a post published by the second object, and may also be other types of text for learning by the user.
In this embodiment, the set of all text compositions published by at least one second object in the second set of objects may be determined to be a join C N 。
Step S106, an interaction result obtained by each first object interacting with each text is obtained.
In the technical scheme provided in the step S106, after at least one text published by each second object in the second object set is obtained, an interaction result obtained by each first object interacting with each text is obtained, wherein in the case that a plurality of first objects exist, interaction results obtained by a plurality of first objects interacting with each text are respectively obtained, a plurality of interaction results are obtained, the interaction results correspond to the first objects one by one, and the interaction result is a quantized result.
Each first object of the embodiment may interact with a specific each text, where the interaction may refer to a domain correlation degree between each first object and each text published by a second object, and the domain correlation degree is also referred to as a domain fitness degree, which may indicate an evaluation authority of each first object on the specific each text published by each second object, and may indicate a expertise (also referred to as an interactive weight) of the text published by the second object in a first object eye in the same domain; the interaction of the embodiment may also be an interaction behavior of each first object to each specific text, for example, a comment, praise, step on, etc.
Step S108, determining authority values of each second object based on the obtained interaction result.
In the technical solution provided in the above step S108 of the present invention, since the interaction result is specific to each text and each text is published by the second object, the interaction result of each first object to each specific text may reflect the authority value of the second object, and both the authority values of the second object may be fed back positively or negatively, so that after the interaction result obtained by each first object interacting with each text is obtained, the authority value of each second object may be determined based on the obtained interaction result.
Through the steps S102 to S108, acquiring a first object set and a second object set, where an authority value of each first object in the first object set reaches a target value; acquiring at least one text published by each second object in the second object set; the method comprises the steps of obtaining an interaction result obtained by each first object interacting with each text; and determining the authority value of each second object based on the obtained interaction result. That is, the authority value of the second object is calculated according to the interaction result of each first object and the second object, even if the second object cannot pass through authentication, the authority value is distinguished in authority, so that the technical problem that the authority value of the object is not effectively determined is solved, the technical effect of effectively determining the authority value of the object is achieved, and the recommendation algorithm of the text is better acted.
The above-described method of this embodiment is further described below.
As an optional implementation manner, step S108, determining the authority value of each second object based on the obtained interaction result, includes: determining authority values of each text based on the obtained interaction result; an authority value for each second object is determined based on the authority value for each text.
In this embodiment, since the interaction result is obtained by each first object interacting with each specific text, the embodiment may first determine the authority value of each text published by each second object based on the obtained interaction result, and then determine the authority value of each second object based on the authority value of each text. Optionally, determining authority values of at least one text published to the second objects according to the method respectively, so as to obtain the authority values of at least one text, and determining the authority value of each second object through the authority values of at least one text.
As an alternative embodiment, determining the authority value of each text based on the obtained interaction result includes: acquiring a first number of interaction results in the obtained interaction results, wherein each interaction result in the first number of interaction results is larger than any interaction result except the first number of interaction results in the obtained interaction results; an authoritative value for each text is determined based on the first number of interaction results.
In this embodiment, when determining the authority value of each text based on the obtained interaction results, since the interaction results are quantized results, in the case that the interaction results are multiple, the multiple interaction results may be ranked from large to small, and then a first number of interaction results that are ranked forward is selected from the multiple interaction results based on the ranked results, where each interaction result in the first number of interaction results is greater than any interaction result in the multiple interaction results except the first number of interaction results. Alternatively, the first number may be 3, which may be adjusted according to the size of the user's volume in the community, and is not a constant value. After the first number of interaction results is obtained, an authoritative value for each corresponding text may be determined based on the first number of interaction results.
As an alternative embodiment, determining the authority value of each text based on the first number of interaction results includes: and determining an average value of the first number of interaction results as an authority value of each text.
In this embodiment, the first number of interaction results may be averaged to determine an authoritative value for the text for which the first number of interaction results are intended.
As an alternative embodiment, determining the authority value of each second object based on the authority value of each text includes: determining a second number of authority values from the authority values of the at least one text, wherein each of the second number of authority values is greater than any one of the authority values of the at least one text other than the second number of authority values; an authority value for each second object is determined based on the second number of authority values.
In this embodiment, the at least one text published by the second object may determine an authority value according to the method, where the at least one text corresponds to the at least one authority value, and where the at least one text corresponds to the plurality of authority values, the plurality of texts may be ranked in order from size, and a second number of authority values with a front ranking result is obtained, where each authority value in the second number of authority values is greater than any authority value except for the second number of authority values in the authority values of the at least one text, the second number may be 10, and may be adjusted according to the size of the user body in the community, and is not a fixed value. After determining the second number of authority values, the authority value of the corresponding second object may be determined based on the second number of authority values.
Any of the second objects in the second set of objects of this embodiment may determine its authority value in accordance with the method described above.
As an alternative embodiment, determining the authority value of each second object based on the second number of authority values comprises: and determining the average value of the authority values of the second number as the authority value of each second object.
In this embodiment, the second number of authority values may be averaged, and the average value thereof may be determined as the authority value of the corresponding second object.
As an optional implementation manner, step S106, obtaining an interaction result obtained by each first object interacting with each text, includes: acquiring the field correlation degree of each first object and each text; the interaction strength of each first object on each text is obtained; and determining an interaction result based on the field relevance and the interaction strength.
In this embodiment, when the interaction result obtained by the interaction of each first object with each text is obtained, the domain relevance between each first object and each text published by the second object may be obtained, where the domain relevance may represent the evaluation authority of each first object on each text, for example, define the domain relevance of a certain article c sent by a certain first object i and a second object u as S ic Representing the authority of the first object to evaluate the article. The higher the field relevance is, the more credible the interaction behavior of the first object to each text published by the second object is.
As an optional implementation manner, obtaining the domain relevance of each first object and each text includes: determining a topic vector for each text; obtaining the similarity between the topic vector of each text and the topic vector of at least one text published by each first object respectively to obtain at least one similarity; the maximum similarity of the at least one similarity is determined as the domain correlation.
In this embodiment, in implementing obtaining the domain relevance between each first object and each text, a topic model of each text specific to the second object publication may be determined, where the topic model may be represented by a topic vector, for example, a topic vector of article c published by the second object u may be represented as T uc . After determining the topic vector of each text, the similarity between the topic vector of each text and the topic vector of at least one text published by each first object can be obtained to obtain at least one similarity, for example, the set of at least one text published by the first object is C Q And obtaining the similarity between the topic vector of the article c and the topic vectors of all texts published by each first object i respectively, and obtaining at least one similarity, wherein the similarity can be the cosine distance between the topic vector of the article c and the topic vector of all texts published by each first object i respectively.
Alternatively, the embodiment may use a document topic generation model (Latent Dirichlet allocation, abbreviated as LDA) algorithm to respectively find a topic vector of each text published by the second object and a topic vector of each text published by the first object.
As an optional implementation manner, obtaining the interaction strength of each first object to each text includes: determining the historical interaction behavior of each first object on each text; and determining the interaction strength according to the weight corresponding to the historical interaction behavior.
In this embodiment, when obtaining the interaction strength of each first object to each text, it may also be determined that each first object performs a historical interaction action on each specific text, where the historical interaction action may be a comment, praise, step on, etc. action performed by each first object on each specific text, and the corresponding weight of the historical interaction action used for determining the interaction strength is provided, for example, the value corresponding to the comment may be 0.8, the value corresponding to the praise may be 0.5, and the step on is 0. Optionally, the weight corresponding to the historical interaction behavior in the embodiment may be adjusted according to the actual situation, and an emotion analysis algorithm may be introduced to calculate positive and negative emotion results for different comment results, so as to determine the interaction strength.
As an alternative embodiment, determining the interaction result based on the domain relevance and the interaction strength includes: and determining the product of the field correlation and the interaction strength as an interaction result.
Acquiring the field correlation degree of each first object and each text by the method; after the interaction strength of each first object to each text is obtained, the product of the domain correlation and the interaction strength can be determined as the final interaction result of each first object to each second object published specific each text.
Optionally, after determining the authority value of each second object based on the obtained interaction result, the method further comprises: and pushing the text published by each second object to the target community based on the authority value of each second object.
In this embodiment, after determining the authority value of each second object based on the obtained interaction result, the authority value of each second object is applied to the text recommendation algorithm, that is, the authority value is considered as a feature of the text recommendation algorithm, the weight of the text recommendation may be determined based on the authority value, alternatively, for the text published by the first object with a higher authority value, the weight may be increased and the recommendation may be performed, and for the text published by the first object with a lower authority value, the weight may be reduced and the recommendation may not be performed. Optionally, the embodiment may determine, from the plurality of second objects, a second object that may push text to the target community based on the authority value of each second object, so as to allow the determined second object to push its published text to the target community, so as to ensure authority of the whole target community.
It should be noted that, in the method, the calculation mode of each part of the method can be adjusted according to the actual situation, and features of time dimension can be added, for example, the influence of the recent text on the finally determined authority value is larger, the influence of the previous text on the authority value is smaller, and the like. The embodiment may also calculate the authority value of the second subject at regular intervals (daily or weekly) to update the authority value.
In this embodiment, based on the first object set with authority values reaching the target value, the authority value of each second object is calculated according to the interaction result of the text published by each first object in the first object set and each second object in the second object set, so that a user who cannot pass authentication has an authority value, and compared with a user who cannot pass authentication, which is considered to have low authority, the authority value of the user cannot pass authentication has a degree of distinction, so that the technical problem that the authority value of the object is not effectively determined is solved, and the technical effect of effectively determining the authority value of the object is achieved, so as to better act on the recommendation algorithm of the text.
Example 2
The technical solution of the embodiment of the present invention is further illustrated in the following description in conjunction with the preferred embodiments, and the text is specifically illustrated as an article.
At present, with the development of the Internet, various network communities with common hobbies or common targets are generated, and people can learn and communicate in the communities. Generally, network communities are divided into two main categories, namely, a community with popular entertainment, most people are entertained in the community, chat is the main, and the other community is a community with preferential learning, most people are learning in the community, and knowledge acquisition is the main. For the former community, the content recommendation algorithm of the community takes the popularity and click rate as the main factors, namely, recommending popular articles with high click rate. In the latter community, since the purpose of the community is to obtain knowledge mainly, the popular articles with high click rate are blogs, but may be nonsensical (hydrologic) from the knowledge perspective, and even contradictory to the fact. For recommendation algorithms, such less authoritative articles should not be given a greater weight in the recommendation.
For the second community, most of recommendation algorithms currently take commercial consideration of user liveness and the like, authority is not considered as a feature of the recommendation algorithm, and less communities can search for more authoritative users in the field, weight is increased for content published by the users and recommendation is performed.
If the factors of the authority of the publisher are not considered in the recommendation algorithm, but only the popularity and click rate of the content are considered, then some articles which are meaningless from the knowledge perspective and even not in fact are given higher recommendation weights, so that the authoritative of the community is harmful to the community in a long term although the community can bring higher liveness in a short term, and the targets of the authors in the community are gradually changed from producing high-quality articles to producing blond-shaped articles, which is also extremely unfavorable to the long-term development of the community.
If the authority assessment user is required to be introduced into the community, a qualification auditing measure is required to be adopted for the user, so that the user submits a relevant qualification certificate, and the community is set as the authority user when the qualification certificate meets a certain condition. However, the authentication strictness of the authoritative users is not well known, if the authentication strictness is too strict, the number of the authoritative users is small, the number of authoritative articles produced by the authoritative users is also small, and the authoritative distinction does not exist for most articles contacted by the common users, so that the essential problem is not solved; if too loose, the number of authoritative users may be excessive, and the articles of those individuals may affect the authority of the entire community.
In order to solve the problem that the authority of the community user cannot be effectively determined, the embodiment provides a community user authority scoring method, the method can enable a user which cannot pass authentication to have an authority value, a part of authoritative seed users can be defined, for each article which is sent by a non-seed user, the field compliance degree of the topic of the non-seed user and the article published by the seed user is calculated, and the higher the compliance degree is, the more credible the interaction behavior of the seed user to the article is indicated. According to the interaction result of the seed user on each article of the non-seed user, positive feedback or negative feedback is carried out on the authority value of the non-seed user, so that the specific authority value of the non-seed user is determined.
The above-described method of this embodiment is described in detail below.
FIG. 2 is a flow chart of a method of determining authority values of another object according to an embodiment of the present invention. As shown in fig. 2, the method may include the steps of:
step S201, an authoritative user set is defined.
The embodiment can adopt a qualification auditing mode to define an authoritative user set, and the users in the authoritative user set are used as seed users. The qualification audit should be rigorous, ensure the authority of the users passing the qualification audit, and preferably cover all knowledge directions of the community. The users in the authoritative user set become authenticated users and can be represented by a set Q, and all articles Cx … … CM published by the users are obtained to obtain an article set C Q 。
Traversing authoritative user Q in authoritative user set Q 1 ……q m Each user uses q j Representing, acquiring article set C published by authoritative user qj The topic vector of all articles published by the authoritative user can be calculated by adopting an LDA algorithm and is recorded as T qj 。
Step S202, obtaining all articles published by users in a non-authentication user set.
Other users in the community except the authoritative user set become non-authenticated users and can be represented by a set N, and the set of all articles published by the users in the non-authenticated user set N can be C N 。
Optionally, traverse C N All articles c 1 ……c n Each article uses c i The representation, this embodiment may employ LDA calculationRespectively obtain C by the method N Subject vector T for each article in (a) ci 。
And step S203, obtaining interaction results of the authoritative user on each article.
This embodiment may define an authority value a (0 < =a < =1) of the user, representing the authority of the user in the community, their authority value a=1 for authenticating the user (seed user).
For interaction of a certain article c sent by an authenticated user i and a non-authenticated user u, the result of the interaction needs to be quantified, and the value consists of two parts: one part is interaction strength (comment is 0.8, praise is 0.5, step on is 0, the value can be adjusted according to actual conditions, even different comments can be calculated by introducing emotion analysis algorithm), the other part is evaluation authority of an authentication user on the article, and the field correlation degree S of an authentication user i and an article c sent by a non-authentication user u can be defined ic The calculation method can be the topic vector T of the article c issued by the non-authenticated user u, which is expressed as the evaluation authority of the authenticated user on the article uc Maximum cosine distance from the topic vector of all articles published by authentication user i.
After the interaction strength and the domain relevance are determined, the interaction result of the authoritative user on each article can be determined from the product between the interaction strength and the domain relevance.
Optionally, in this embodiment, all authoritative users Q in the set of authoritative users Q are traversed 1 ……q m For each authoritative user q j If q j And article c i Without interaction, the authoritative user has an interaction value a for the article ciqj Is 0; if authoritative user q j And article c i If there is interaction, the interaction value a of the authoritative user to the article ciqj The method comprises the following steps:
cos < Tci, tqj >. W, where w has a value of (comment 0.8, praise 0.5, step on 0, also can be adjusted according to the actual situation or emotion analysis added to the comment).
And step S204, based on the interaction result of the authoritative user on each article, determining the authoritative value of each article.
For an article published by a non-authenticated user, the authority of the article is the average of the highest three values (tunable) of the quantized values of the interaction results of all authenticated users under the article. For example, for [ a ] ciq1 ,……,a ciqm ]Ordering from big to small, taking the first three numbers, which are assumed to be [ a ] ci0 ,a ci1 ,a ci2 ]The average value is taken and is determined as the authority value of each article.
Step S205, the authority value of the non-authenticated user is determined based on the authority value of each article.
The embodiment may determine an average of the highest 10 values (tunable) of the article authority values of all articles issued by the non-authenticated user as the authority value of the non-authenticated user. Authoritative value for all articles [ a ] c1 ,a c2 ,……,a cn ]Sorting from big to small, taking the largest 10 numbers, averaging the numbers, and determining the average value as the authority value of each article and the authority value of the non-authenticated user.
This embodiment may calculate the authority value for updating the user daily or weekly.
It should be noted that, in the embodiment, the weight can be adjusted by itself in different interactions (such as point and stepping), and other algorithms (such as emotion analysis algorithm analysis comment positive and negative emotion) can be added to obtain the weight; the method is used for calculating 10 values reserved by the authority values of the users and 3 values reserved by the authority values of the articles, and the reserved quantity can be specifically adjusted according to the size of the user body in the community and is not a constant value.
The embodiment provides a community user authority scoring method, which is a method for calculating authority values of common users according to interaction results of the authority users and the common users, and is characterized in that the expertise (interaction weight) of the common users in eyes of experts in the same field (cosine distance similarity between articles published by the field experts and subject models posted by the users) is calculated, and the authority of the common users is calculated according to historical interaction behaviors.
It should be noted that, the calculation mode of each part of the method can be adjusted according to the actual situation, and features of time dimension can be added, for example, the influence of the latest article on the result is larger, the influence of the former article on the result is smaller, and the like.
Example 3
The embodiment of the invention also provides a device for determining the authority value of the object. It should be noted that the apparatus for determining an authority value of an object according to this embodiment may be used to perform the method for determining an authority value of an object according to the embodiment of the present invention.
FIG. 3 is a schematic diagram of an apparatus for determining authority values of an object according to an embodiment of the present invention. As shown in fig. 3, the determining means 30 of the authority value of the object may include: a first acquisition unit 31, a second acquisition unit 32, a third acquisition unit 33, and a determination unit 34.
A first obtaining unit 31, configured to obtain a first object set and a second object set, where an authority value of each first object in the first object set reaches a target value.
A second obtaining unit 32, configured to obtain at least one text published by each second object in the second object set.
And a third obtaining unit 33, configured to obtain an interaction result obtained by each first object interacting with each text.
A determining unit 34, configured to determine an authority value of each second object based on the obtained interaction result.
Optionally, the determining unit 34 includes: the first determining module is used for determining authority values of each text based on the obtained interaction result; and the second determining module is used for determining the authority value of each second object based on the authority value of each text.
Optionally, the first determining module includes: the first acquisition sub-module is used for acquiring a first number of interaction results in the obtained interaction results, wherein each interaction result in the first number of interaction results is larger than any interaction result except the first number of interaction results in the obtained interaction results; the first determining sub-module is used for determining authority values of each text based on the first number of interaction results.
Optionally, the determining submodule is configured to determine the authority value of each text based on the first number of interaction results by: and determining an average value of the first number of interaction results as an authority value of each text.
Optionally, the second determining module includes: a second determining sub-module for determining a second number of authority values from the authority values of the at least one text, wherein each of the second number of authority values is greater than any one of the authority values of the at least one text other than the second number of authority values; a third determination submodule for determining the authority value of each second object based on the authority value of the second number.
Optionally, the third determining submodule is configured to determine an authority value of each second object based on the second number of authority values by: and determining the average value of the authority values of the second number as the authority value of each second object.
Optionally, the third acquisition unit 33 includes: the first acquisition module is used for acquiring the field correlation degree of each first object and each text; the second acquisition module is used for acquiring the interaction strength of each first object on each text; and the third determining module is used for determining an interaction result based on the field relevance and the interaction strength.
Optionally, the first acquisition module includes: a fourth determination submodule for determining a topic vector of each text; the second obtaining submodule is used for obtaining the similarity between the topic vector of each text and the topic vector of at least one text published by each first object respectively to obtain at least one similarity; and a fifth determining sub-module for determining the maximum similarity of the at least one similarity as the domain correlation.
Optionally, the second acquisition module includes: a sixth determining submodule, configured to determine a historical interaction behavior of each first object on each text; and the seventh determining submodule is used for determining the interaction strength according to the weight corresponding to the historical interaction behavior.
Optionally, the third determining module includes: and the eighth determining submodule is used for determining the product of the field relativity and the interaction strength as an interaction result.
Optionally, the apparatus further comprises: and the pushing unit is used for pushing the text published by each second object to the target community based on the authority value of each second object after the authority value of each second object is determined based on the obtained interaction result.
In this embodiment, the first object set and the second object set are acquired by the first acquisition unit 31, wherein the authority value of each first object in the first object set reaches the target value; acquiring at least one text published by each second object in the second object set through a second acquiring unit 32; the interaction result obtained by the interaction of each first object with each text is acquired through the third acquisition unit 33; by determining the authority value of each second object based on the obtained interaction result by the determining unit 34, the technical problem that the authority value of the object is not effectively determined is solved, and the technical effect of effectively determining the authority value of the object is achieved.
Example 4
According to an embodiment of the present invention, there is also provided a storage medium including a stored program, wherein the program performs the method of determining an authority value of an object described in embodiment 1.
Example 5
According to an embodiment of the present invention, there is further provided a processor for running a program, where the program runs to perform the method for determining an authority value of an object described in embodiment 1.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (12)
1. A method for determining an authority value of an object, comprising:
acquiring a first object set and a second object set, wherein the authority value of each first object in the first object set reaches a target value;
acquiring at least one text published by each second object in the second object set;
the interaction result obtained by each first object interacting with each text is obtained;
determining authority values of each second object based on the obtained interaction results;
the method for obtaining the interaction result obtained by each first object interacting with each text comprises the following steps: acquiring the field correlation degree of each first object and each text; acquiring interaction strength of each first object on each text; determining the interaction result based on the domain relevance and the interaction strength;
Determining an authority value of each second object based on the obtained interaction result, including: determining authority values of each text based on the obtained interaction result; an authoritative value for each of the second objects is determined based on the authoritative value for each of the texts.
2. The method of claim 1, wherein determining an authoritative value for each of the texts based on the obtained interaction results comprises:
acquiring a first number of interaction results in the obtained interaction results, wherein each interaction result in the first number of interaction results is larger than any interaction result except the first number of interaction results in the obtained interaction results;
an authoritative value for each of the texts is determined based on the first number of interaction results.
3. The method of claim 2, wherein determining an authoritative value for each of the texts based on the first number of interaction results comprises:
and determining an average value of the first number of interaction results as an authority value of each text.
4. The method of claim 1, wherein determining an authority value for each of the second objects based on the authority value for each of the texts comprises:
Determining a second number of authority values from the authority values of the at least one text, wherein each of the second number of authority values is greater than any one of the authority values of the at least one text other than the second number of authority values;
an authority value for each of the second objects is determined based on the second number of authority values.
5. The method of claim 4, wherein determining an authority value for each of the second objects based on the second number of authority values comprises:
and determining an average value of the authority values of the second quantity as the authority value of each second object.
6. The method of claim 1, wherein obtaining a domain relevance of each of the first objects to each of the texts comprises:
determining a topic vector for each of the texts;
obtaining the similarity between the topic vector of each text and the topic vector of at least one text published by each first object respectively to obtain at least one similarity;
and determining the maximum similarity in the at least one similarity as the field relevance.
7. The method of claim 1, wherein obtaining the interaction strength of each first object with each text comprises:
Determining a historical interaction behavior of each first object on each text;
and determining the interaction strength according to the weight corresponding to the historical interaction behavior.
8. The method of claim 1, wherein determining the interaction result based on the domain relevance and the interaction strength comprises:
and determining the product of the field correlation degree and the interaction strength as the interaction result.
9. The method according to any one of claims 1 to 8, wherein after determining an authority value of each of the second objects based on the obtained interaction result, the method further comprises:
and pushing the text published by each second object to a target community based on the authority value of each second object.
10. An apparatus for determining an authority value of an object, comprising:
the first acquisition unit is used for acquiring a first object set and a second object set, wherein the authority value of each first object in the first object set reaches a target value;
the second obtaining unit is used for obtaining at least one text published by each second object in the second object set;
The third acquisition unit is used for acquiring an interaction result obtained by each first object interacting with each text;
the determining unit is used for determining authority values of each second object based on the obtained interaction result;
wherein the third acquisition unit includes: the first acquisition module is used for acquiring the field correlation degree between each first object and each text; the second acquisition module is used for acquiring interaction strength of each first object on each text; the determining module is used for determining the interaction result based on the field relevance and the interaction strength;
the determination unit includes: the first determining module is used for determining authority values of each text based on the obtained interaction result; and the second determining module is used for determining the authority value of each second object based on the authority value of each text.
11. A storage medium comprising a stored program, wherein the program, when run, controls a device in which the storage medium is located to perform the method of any one of claims 1 to 9.
12. A processor for running a program, wherein the program when run performs the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010296679.9A CN111523049B (en) | 2020-04-15 | 2020-04-15 | Method, device, storage medium and processor for determining authority value of object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010296679.9A CN111523049B (en) | 2020-04-15 | 2020-04-15 | Method, device, storage medium and processor for determining authority value of object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111523049A CN111523049A (en) | 2020-08-11 |
| CN111523049B true CN111523049B (en) | 2023-06-13 |
Family
ID=71903065
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010296679.9A Active CN111523049B (en) | 2020-04-15 | 2020-04-15 | Method, device, storage medium and processor for determining authority value of object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111523049B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116660A (en) * | 2013-03-15 | 2013-05-22 | 人民搜索网络股份公司 | Method and device for acquiring website authority values |
| CN103136330A (en) * | 2013-01-04 | 2013-06-05 | 武汉大学 | User reliability assessment method based on microblog platforms |
| CN104317881A (en) * | 2014-04-11 | 2015-01-28 | 北京理工大学 | Method for reordering microblogs on basis of authorities of users' topics |
| CN104838376A (en) * | 2012-10-04 | 2015-08-12 | 谷歌公司 | Generating snippets for prominent users for information retrieval queries |
| CN107169873A (en) * | 2017-06-16 | 2017-09-15 | 北京信息科技大学 | A kind of microblog users authority evaluation method of multiple features fusion |
| CN107577827A (en) * | 2017-10-30 | 2018-01-12 | 北京国舜科技股份有限公司 | A kind of method and device of internet information grading |
-
2020
- 2020-04-15 CN CN202010296679.9A patent/CN111523049B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104838376A (en) * | 2012-10-04 | 2015-08-12 | 谷歌公司 | Generating snippets for prominent users for information retrieval queries |
| CN103136330A (en) * | 2013-01-04 | 2013-06-05 | 武汉大学 | User reliability assessment method based on microblog platforms |
| CN103116660A (en) * | 2013-03-15 | 2013-05-22 | 人民搜索网络股份公司 | Method and device for acquiring website authority values |
| CN104317881A (en) * | 2014-04-11 | 2015-01-28 | 北京理工大学 | Method for reordering microblogs on basis of authorities of users' topics |
| CN107169873A (en) * | 2017-06-16 | 2017-09-15 | 北京信息科技大学 | A kind of microblog users authority evaluation method of multiple features fusion |
| CN107577827A (en) * | 2017-10-30 | 2018-01-12 | 北京国舜科技股份有限公司 | A kind of method and device of internet information grading |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111523049A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2016247184B2 (en) | Attribute weighting for media content-based recommendation | |
| Kumar et al. | Deep Neural Architecture for News Recommendation. | |
| CN107341268B (en) | Hot searching ranking method and system | |
| Wakil et al. | Improving web movie recommender system based on emotions | |
| CN107301247B (en) | Method and device for establishing click rate estimation model, terminal and storage medium | |
| EP4181026A1 (en) | Recommendation model training method and apparatus, recommendation method and apparatus, and computer-readable medium | |
| CN110334356B (en) | Article quality determining method, article screening method and corresponding device | |
| CN106407364B (en) | Information recommendation method and device based on artificial intelligence | |
| CN111767429A (en) | Video recommendation method and device and electronic equipment | |
| CN113407854A (en) | Application recommendation method, device and equipment and computer readable storage medium | |
| Modani et al. | Fairness aware recommendations on behance | |
| EP2613275B1 (en) | Search device, search method, search program, and computer-readable memory medium for recording search program | |
| CN111881365A (en) | Content recommendation method and device | |
| Yu et al. | Exploring folksonomy and cooking procedures to boost cooking recipe recommendation | |
| Meidl et al. | Using game reviews to recommend games | |
| CN111523049B (en) | Method, device, storage medium and processor for determining authority value of object | |
| US20160132771A1 (en) | Application Complexity Computation | |
| CN106469163A (en) | A kind of public number recommends method and system | |
| Marini et al. | There is more to attraction than meets the eye: Studying decoy‐induced attention allocation without eye tracking | |
| Desarkar et al. | Diversification in news recommendation for privacy concerned users | |
| US20170372170A1 (en) | Network-based content submission and contest management | |
| Schmidt et al. | Polarization rank: a study on European News Consumption on Facebook | |
| CN109086813A (en) | Determination method, apparatus, equipment and the storage medium of main broadcaster's similarity | |
| CN106294487B (en) | Self-adapted search method, equipment and system Internet-based | |
| CN113468394A (en) | Data processing method and device, electronic equipment and storage medium |
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 |