CN109886702A - The method and apparatus of abnormal behaviour in a kind of judgement business activity - Google Patents
The method and apparatus of abnormal behaviour in a kind of judgement business activity Download PDFInfo
- Publication number
- CN109886702A CN109886702A CN201711260817.2A CN201711260817A CN109886702A CN 109886702 A CN109886702 A CN 109886702A CN 201711260817 A CN201711260817 A CN 201711260817A CN 109886702 A CN109886702 A CN 109886702A
- Authority
- CN
- China
- Prior art keywords
- user
- evaluation information
- threshold
- emotional
- evaluation
- 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.)
- Pending
Links
Abstract
The invention discloses a kind of method and apparatus of abnormal behaviour in judgement business activity, are related to field of computer technology.One specific embodiment of this method includes: to belong to the user of same user community according to the incidence relation identification between user；The evaluation information of user's publication in each user community is analyzed, with the tendentiousness of the determination evaluation information；According to abnormal behaviour of the user community in the business activity described in the tendency sex determination of the evaluation information of user each in user community publication.The embodiment does not need to can be improved the efficiency for determining abnormal behaviour in business activity by artificially reporting.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of methods and dress for determining abnormal behaviour in business activity
It sets.
Background technique
Currently, some businessmans can be in sale (such as merchandising or service) using the single abnormal row of brush for number one
For for example, it is the behavior of the commodity that businessman provides or service brush evaluation, the behavior that businessman, which employs batch brush (Ji Shuadan group),
It is divided into two classes, one kind is brushed and commented for the commodity of oneself or service, and two classes are commented for the commodity or service brush difference of rival, should
Behavior has misled the purchase intention of user significantly.
Identification abnormal behaviour punishes the businessman there are abnormal behaviour mostly or by artificial report at present.
In realizing process of the present invention, at least there are the following problems in the prior art for inventor's discovery:
There are problems that low efficiency by way of artificially reporting and determining abnormal behaviour.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of method and apparatus for determining abnormal behaviour in business activity, it is not required to
It can be improved the efficiency for determining abnormal behaviour in business activity by artificially reporting.
To achieve the above object, according to an aspect of an embodiment of the present invention, it provides different in a kind of judgement business activity
The method of Chang Hangwei.
A kind of method of abnormal behaviour in judgement business activity, comprising: belonged to according to the incidence relation identification between user same
The user of one user community；The evaluation information of user's publication in each user community is analyzed, with the determination evaluation letter
The tendentiousness of breath；According to user community described in the tendency sex determination of the evaluation information of user each in user community publication in institute
State the abnormal behaviour in business activity.
Optionally, the step of user of same user community being belonged to according to the incidence relation identification between user, comprising: according to
Community structure discovery algorithm determines the incidence relation between the user；It deletes and is less than or equal to default threshold with business subject evaluation rate
Incidence relation between two users of value；After the operation for completing the deletion, according to the current incidence relation of the user
Identification belongs to the user of same user community；Wherein, the same business subject evaluation rate are as follows: two users are to identical services object
The sum of evaluation quantity with two users to the ratio of all evaluation quantity summations of business object.
Optionally, the evaluation information of user's publication in each user community is analyzed, with the determination evaluation information
Tendentious step, comprising: in each user community user publication evaluation information carry out emotional semantic analysis；According to institute
The result for stating emotional semantic analysis calculates the emotional value of every evaluation information；Every evaluation letter is determined according to the emotional value
The tendentiousness of breath.
Optionally, the step of emotional semantic analysis being carried out to the evaluation information of user's publication in each user community, comprising:
It is segmented according to the evaluation information that preset segmentation methods issue user each in each user community；According to preset emotion
It analyzes dictionary and emotional semantic analysis is carried out to the participle, to obtain the negative word in the participle, degree word, emotion word difference
Corresponding weighted value.
Optionally, every evaluation information includes one or more sense-groups, the result analyzed according to the emotional semantic
The step of calculating the emotional value of every evaluation information, comprising: according to the negative word, degree word, the corresponding power of emotion word
Weight values calculate the sense-group emotional value of each sense-group；According to each sense-group emotional value in every evaluation information it
With obtain the emotional value of every evaluation information.
Optionally, the tendentious step of every evaluation information is determined according to the emotional value, comprising: will be described every
The emotional value of evaluation information is compared with first threshold, second threshold, and the first threshold is less than the second threshold, such as
The emotional value of one evaluation information of fruit is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；If one comments
The emotional value of valence information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is favorable comment；If an evaluation information
The emotional value be greater than first threshold and be less than second threshold, it is determined that the tendentiousness of the evaluation information be neutrality.
According to another aspect of an embodiment of the present invention, a kind of device for determining abnormal behaviour in business activity is provided.
The device of abnormal behaviour in a kind of judgement business activity, comprising: identification module, for being closed according to the association between user
System's identification belongs to the user of same user community；Analysis module, for the evaluation information to user's publication in each user community
It is analyzed, with the tendentiousness of the determination evaluation information；Determination module, for being issued according to user each in the user community
Evaluation information tendency sex determination described in abnormal behaviour of the user community in the business activity.
Optionally, the identification module is also used to: determining that the association between the user is closed according to community structure discovery algorithm
System；Delete the incidence relation being less than or equal between two users of preset threshold with business subject evaluation rate；It deletes described in the completion
After the operation removed, the user of same user community is belonged to according to the current incidence relation identification of the user；Wherein, described same
Business object Assessment Rate are as follows: two users are to two users of the sum of evaluation quantity of identical services object and this to business object
The ratio of all evaluation quantity summations.
Optionally, the analysis module is also used to: carrying out emotion to the evaluation information of user's publication in each user community
Semantic analysis；The emotional value of every evaluation information is calculated according to the result that the emotional semantic is analyzed；It is true according to the emotional value
The tendentiousness of fixed every evaluation information.
Optionally, the analysis module includes emotional semantic analysis module, and the emotional semantic analysis module is used for: according to
The evaluation information that preset segmentation methods issue user each in each user community segments；According to preset sentiment analysis
Dictionary carries out emotional semantic analysis to the participle, is respectively corresponded with obtaining the negative word in the participle, degree word, emotion word
Weighted value.
Optionally, every evaluation information includes one or more sense-groups, and the analysis module further includes emotional value meter
Calculate module, the emotional value computing module by: based on the negative word, degree word, the corresponding weighted value of emotion word
Calculate the sense-group emotional value of each sense-group；According to the sum of each sense-group emotional value in every evaluation information, obtain
The emotional value of every evaluation information.
Optionally, the analysis module includes tendentiousness determining module, and the tendentiousness determining module is used for: will be described every
The emotional value of evaluation information is compared with first threshold, second threshold, and the first threshold is less than the second threshold, such as
The emotional value of one evaluation information of fruit is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；If one comments
The emotional value of valence information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is favorable comment；If an evaluation information
The emotional value be greater than first threshold and be less than second threshold, it is determined that the tendentiousness of the evaluation information be neutrality.
Another aspect according to an embodiment of the present invention, provides a kind of electronic equipment.
A kind of electronic equipment, comprising: one or more processors；Memory works as institute for storing one or more programs
When stating one or more programs and being executed by one or more of processors, determine so that one or more of processors are realized
The method of abnormal behaviour in business activity.
Another aspect according to an embodiment of the present invention, provides a kind of computer-readable medium.
A kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
The method for determining abnormal behaviour in business activity is realized when row.
One embodiment in foregoing invention has the following advantages that or the utility model has the advantages that is identified according to the incidence relation between user
Belong to the user of same user community；The evaluation information of user's publication in each user community is analyzed, to determine evaluation
The tendentiousness of information；When the tendentiousness of the evaluation information of user each in user community publication be consistent favorable comment or it is consistent it is poor comment,
Then determine that there are abnormal behaviours in business activity for the user community.So that not needing to can be improved judgement by artificially reporting
The efficiency of abnormal behaviour in business activity.
Further effect possessed by above-mentioned non-usual optional way adds hereinafter in conjunction with specific embodiment
With explanation.
Detailed description of the invention
Attached drawing for a better understanding of the present invention, does not constitute an undue limitation on the present invention.Wherein:
Fig. 1 is the key step schematic diagram of the method for abnormal behaviour in judgement business activity according to an embodiment of the present invention；
Fig. 2 is group's brush single act of electric business platform businessman in judgement merchandise sales business according to an embodiment of the present invention
The preferred flow schematic diagram of method；
Fig. 3 is the main modular schematic diagram of the device of abnormal behaviour in judgement business activity according to an embodiment of the present invention；
Fig. 4 is that the embodiment of the present invention can be applied to exemplary system architecture figure therein；
Fig. 5 is adapted for the structural schematic diagram for the computer system for realizing the server of the embodiment of the present invention.
Specific embodiment
Below in conjunction with attached drawing, an exemplary embodiment of the present invention will be described, including the various of the embodiment of the present invention
Details should think them only exemplary to help understanding.Therefore, those of ordinary skill in the art should recognize
It arrives, it can be with various changes and modifications are made to the embodiments described herein, without departing from scope and spirit of the present invention.Together
Sample, for clarity and conciseness, descriptions of well-known functions and structures are omitted from the following description.
Fig. 1 is the key step schematic diagram of the method for abnormal behaviour in judgement business activity according to an embodiment of the present invention.
Business activity can be the commodity or service sales business of electric business platform, distribution via internet, network food delivery etc..
As shown in Figure 1, the method for abnormal behaviour mainly includes following step in the judgement business activity of the embodiment of the present invention
Rapid S101 to step S103.
Step S101: belong to the user of same user community according to the incidence relation identification between user.
It the step of belonging to the user of same user community according to the incidence relation identification between user, can specifically include:
The incidence relation between user is determined according to community structure discovery algorithm；
Delete the incidence relation being less than or equal between two users of preset threshold with business subject evaluation rate；
After the operation for completing the deletion, the use of same user community is belonged to according to the current incidence relation identification of user
Family；
It wherein, can be with business subject evaluation rate are as follows: two users are to the sum of evaluation quantity of identical services object and this
Ratio of two users to all evaluation quantity summations of business object.
It is associated between two users by evaluating identical one or more business objects, business object specifically may be used
Think commodity or service, the quantity for the evaluation information that evaluation quantity, that is, user issues the business objects such as commodity or service.With industry
For object be engaged in as commodity of sale such as electric business businessmans, same commodity Assessment Rate mutually should be with business subject evaluation rate.It is commented with commodity
Valence rate is for two users to two users of the sum of evaluation quantity of identical commodity and this to all evaluation quantity summations of each commodity
Ratio.
Step S102: the evaluation information of user's publication in each user community is analyzed, to determine evaluation information
Tendentiousness.
The evaluation information of user's publication in each user community is analyzed, with the tendentiousness of the determination evaluation information
The step of, it can specifically include:
Emotional semantic analysis is carried out to the evaluation information of user's publication in each user community；
The emotional value of every evaluation information is calculated according to the result that emotional semantic is analyzed；
The tendentiousness of every evaluation information is determined according to emotional value.
Wherein, the step of carrying out emotional semantic analysis to the evaluation information of user's publication in each user community, specifically may be used
To include:
It is segmented according to the evaluation information that preset segmentation methods issue user each in each user community；
Emotional semantic analysis is carried out to participle according to preset sentiment analysis dictionary, with the negative word in being segmented, journey
Spend word, the corresponding weighted value of emotion word.
Every evaluation information may include one or more sense-groups.
The step of calculating the emotional value of every evaluation information according to the result that emotional semantic is analyzed, specifically includes:
The sense-group emotional value of each sense-group is calculated according to negative word, degree word, the corresponding weighted value of emotion word, specifically
Ground, sense-group emotional value=negative word weight * degree word weight * emotion word weight；
According to the sum of each sense-group emotional value in every evaluation information, the emotional value of every evaluation information is obtained, specifically,
The emotional value of one evaluation information are as follows: sense-group emotional value 1+ sense-group emotional value 2 ,+...+sense-group emotional value n, wherein n is sense-group
Quantity, sense-group refer to that each ingredient marked off in sentence by the meaning and structure, each ingredient are known as a sense-group, can be by one
Multiple ingredients made of being divided in evaluation information by punctuation mark (such as comma, branch etc.) are as this evaluation information
Multiple sense-groups.
Tendentiousness may include favorable comment, difference is commented and neutrality.The tendentious of every evaluation information is determined according to the emotional value
Step can specifically include:
The emotional value of every evaluation information is compared with first threshold, second threshold, and first threshold is less than
Two threshold values,
If the emotional value of an evaluation information is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；
If the emotional value of an evaluation information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is favorable comment；
If the emotional value of an evaluation information is greater than first threshold and is less than second threshold, it is determined that the evaluation information inclines
Tropism is neutrality.
Step S103: when the tendentiousness of the evaluation information of user each in user community publication is consistent favorable comment or consistent poor
It comments, then determines that there are abnormal behaviours in business activity for the user community.
When user each in a user community publication evaluation information tendentiousness be favorable comment (i.e. consistent favorable comment) or
(i.e. unanimously poor to comment) is commented for difference, then determines that there are abnormal behaviours in business activity for the user community.
The method of abnormal behaviour can be used for determining the business such as business, sale in the judgement business activity of the embodiment of the present invention
Abnormal behaviour present in activity, specifically such as determine electric business platform, distribution via internet, businessman network food delivery in merchandising or
The abnormal behaviours such as group's brush list in the business activity of service, brush is single i.e. by evaluating commodity brush list (brush evaluation information), specifically
It may include brushing to comment or brush difference is commented.
It is brushed below by the group of the merchandise sales business of the businessman of electric business platform, judgement electric business platform businessman of business activity
For single act, the method for determining abnormal behaviour in business activity of the embodiment of the present invention is discussed in detail.
The method of group's brush single act of electric business platform businessman passes through in the judgement merchandise sales business of the embodiment of the present invention
GN algorithm (a kind of community structure discovery algorithm) finds one or more user community relatively high with commodity Assessment Rate, these
Simultaneously multiple and different commodity are carried out with the evaluation of commodity dimension with the relatively high user community of commodity Assessment Rate, and evaluate
Commodity largely be all it is identical, that is, show that the same commodity Assessment Rate of these user communities is relatively high, for same commodity Assessment Rate
A large amount of evaluation informations of the publication of one or more relatively high user community carry out the emotional semantic analysis of commodity dimension, i.e.,
Emotional semantic analysis is carried out to a large amount of evaluation informations of the same user community to multiple commodity, if high with commodity Assessment Rate
The user community of (being higher than a preset threshold) is all consistent favorable comment to the evaluation information of multiple commodity or difference is commented, then judges this
There are group's brush single acts for class user community, if with the high user community of commodity Assessment Rate for the same commodity evaluation both
There is favorable comment to there is difference to comment again, is then judged as normal user community, that is, the single act of group's brush is not present.Wherein, judge user community
Whether favorable comment is all consistent to the evaluation information of multiple commodity or difference is commented can set a judgment criteria, specific judgement mark
Brigadier is described in detail below.
Fig. 2 is group's brush single act of electric business platform businessman in judgement merchandise sales business according to an embodiment of the present invention
The preferred flow schematic diagram of method.
As shown in Fig. 2, the group's brush single act for determining electric business platform businessman in merchandise sales business of the embodiment of the present invention
The preferred flow of method include the following steps, namely S201 to step S208.
Step S201: the incidence relation between user is determined according to GN algorithm.
GN algorithm is a kind of community structure discovery algorithm of Schizoid.The algorithm according to high cohesion inside community in network,
Between community the characteristics of low cohesion, intercommunal side is gradually removed, obtains the community structure of opposite cohesion.GN algorithm Bian Jie
Several concepts detects the position on side, certain while while betweenness be defined as the shortest path on network between all vertex and pass through the side
Number.By definition it is found that the shortest path between the two community's nodes passes through if a line connects the community Liang Ge
The number on the side will be most, and corresponding side betweenness is maximum.If deleting the side, Liang Ge corporations will be separated.GN is calculated
Method is namely based on the shortest path that this thought calculates current network repeatedly, calculates the side betweenness of each edge, and it is maximum to delete side betweenness
Side, the community structure of network can be obtained in algorithm after stopping.
It is associated between every two user by evaluating same or multiple commodity in the embodiment of the present invention, thus institute
Have user because between incidence relation and form a network, each user is equivalent to community's node in network,
It is established and is associated with by side between community's node, pass through the community structure of the above-mentioned available network of GN algorithm, the society of the network
Plot structure reflects the incidence relation between each user.
Step S202: the incidence relation being less than or equal between two users of preset threshold with commodity Assessment Rate is deleted.
Wherein, with commodity Assessment Rate (SPRi,j) can be calculated with following formula:
Wherein, SPCi,jIt is user i and user j to the sum of the evaluation quantity of identical commodity, APCi,jFor user i and user j
To all evaluation quantity summations of each commodity, the quantity that quantity is the evaluation information issued is evaluated.
Step S203: belong to the user of same user community according to the incidence relation identification of each user's current residual.
In the network being made of each user, same commodity Assessment Rate is less than or equal between two users of preset threshold
It, can will there are still the incidence relation (passes of current residual there are still incidence relation between certain user after incidence relation is deleted
Connection relationship) user's identification belong to the user of same user community, to obtain one or more user communities, these user groups
The same commodity Assessment Rate of every two user is above preset threshold in body, and indicating these user communities, there are the possibility of brush single act
Property it is larger, may determine that these user communities with the presence or absence of brush single act by further analyzing.
Step S202, the step S203 of the embodiment of the present invention are executed when realizing GN algorithm, specifically, are preset
The stop condition of GN algorithm are as follows: when the same commodity Assessment Rate between every two user is all higher than preset threshold, algorithm stops, that
, according to the value of same commodity Assessment Rate, preset threshold is less than or equal to according to the value that GN algorithm deletes same commodity Assessment Rate one by one
Two nodes between side, that is, by same commodity Assessment Rate be less than or equal to threshold value two users between incidence relation divide
It opens, to the last among all nodes (i.e. user) in network, the same commodity Assessment Rate between every two user is all higher than pre-
If when threshold value, GN algorithm stops, the incidence relation of user's current residual in network is obtained, according still further to the pass of each user's current residual
Connection relation recognition belongs to the user of same user community.
Step S204: emotional semantic analysis is carried out to the evaluation information of user's publication in each user community.
It can use machine learning method and emotion language carried out to a large amount of evaluation informations of the user belonged in each user community
Justice analysis.
The process for carrying out emotional semantic analysis to the evaluation information of user's publication in each user community particularly may be divided into text
This cutting and emotion position two processes.
Text cutting carries out the evaluation information that user each in each user community issues according to preset segmentation methods
Participle.Wherein, preset segmentation methods can use various segmentation methods, and the jieba participle (stammerer of open source can be used for example
Participle) algorithm segments the evaluation information of commodity to what each user issued, and which can support three kinds of participle moulds
Formula:
1, accurate model, the mode attempt most accurately to cut sentence, are suitble to text analyzing.Participle mode is for example: I/
Come/Beijing/Tsinghua University.
2, syntype, the mode can all scan all in sentence at the word of word, feature be speed very
Fastly, but it not can solve and the problem of ambiguity occur.Participle mode is for example: I/come/Beijing/Tsing-Hua University/Tsinghua University/Hua Da/big
It learns.
3, search engine mode, the mode to long word cutting again, improve recall rate, fit on the basis of accurate model
It shares and is segmented in search engine.Participle mode is for example: I/come/Beijing/Tsing-Hua University/Hua Da/university/Tsinghua University.
Jieba participle (stammerer participle) algorithm of the embodiment of the present invention can be segmented using syntype, its main feature is that algorithm
Time overhead is small, it should be noted that accurate model and search engine mode are also suitable the embodiment of the present invention.
Emotion positioning carries out emotional semantic analysis to the participle according to preset sentiment analysis dictionary, described to obtain
Negative word, degree word, the corresponding weighted value of emotion word in participle.It specifically, can be based on the Chinese emotion prestored point
Analyse dictionary, in advance construct an emotion vocabulary, text cutting process execute word segmentation processing after, by the word obtained after processing according to
Secondary to search one by one in the emotion vocabulary built in advance, if some word can be found, which is emotion word, and is read
Feeling polarities and corresponding weight value, otherwise, the word are not emotion words, then execute the lookup to next word, until whole sentence is commented
The judgement of valence information terminates.Feeling polarities can be positive or negative, and feeling polarities are that the weight of the affirmative emotion word of affirmative is, for example,
1, feeling polarities are that the weight of the negative emotion word of negative is, for example, -1.Chinese sentiment analysis dictionary for example can be using emotion point
Analysis further includes Chinese degree rank word (degree word) totally 219 in the Chinese sentiment analysis dictionary with word collection (beta editions),
And degree word (i.e. degree adverb) is divided six grades, a weight is defined for each degree adverb, for example, the power of " very "
Value is defined as 2.Emotion word its weight after being modified by degree adverb should adjust accordingly, and by the weight definition of negative word
It is -1.
Step S205: the emotional value of every evaluation information is calculated according to the result that emotional semantic is analyzed.
Every evaluation information includes one or more sense-groups, such as: " I thinks that this commodity is very bad, simply too poor
", wherein " I thinks that this commodity is very bad " is a sense-group, and " simply too poor " is another sense-group.
It specifically can be according to emotional semantic according to the emotional value that the result that emotional semantic is analyzed calculates every evaluation information
Analysis calculates often as a result, analyzing the corresponding weighted value of the negative word obtained, degree word, emotion word by emotional semantic
The sense-group emotional value of each sense-group of evaluation information, can calculate according to the following formula:
Sense-group emotional value=negative word weight * degree word weight * emotion word weight
Such as: " I thinks that this commodity is very bad, simply too poor ", wherein " very " being degree pair in first sense-group
Word, " no " are negative word, and it " too " is degree adverb in second sense-group that " good ", which is emotion word, " poor " to negate emotion word, can be with
Be interpreted as " bad ", then, the sense-group emotional value of first sense-group i.e.:
Sense-group emotional value 1=-1 (no) * 2 (very) * 1 (good)=- 2；
The sense-group emotional value of second sense-group is i.e.:
Sense-group emotional value 2=2 (too) * -1 (poor)=- 2,
According to the sum of each sense-group emotional value in every evaluation information, the emotional value of every evaluation information can be obtained.Accordingly
The formula that ground calculates the emotional value of an evaluation information is as follows:
SUM (sense-group emotional value 1, sense-group emotional value 2 ... sense-group emotional value n),
Wherein, n is the quantity of sense-group.
So, the emotional value of " I thinks that this commodity is very bad, simply too poor " this evaluation information is in upper example are as follows:
SUM (sense-group emotional value 1, sense-group emotional value 2)=(- 2)+(- 2)=- 4
Step S206: the tendentiousness of every evaluation information is determined according to the emotional value of every evaluation information.
The tendentiousness of every evaluation information specifically includes favorable comment, difference is commented and neutral.
Two threshold values: first threshold and second threshold can be preset, and first threshold is less than the second threshold.It is logical
The emotional value for crossing every evaluation information is compared with first threshold, second threshold respectively, if the emotion of an evaluation information
Value is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；If the emotional value of one article of evaluation information is greater than the
Two threshold values, it is determined that the tendentiousness of the evaluation information is favorable comment；If the emotional value of an evaluation information be greater than first threshold and
Less than second threshold, it is determined that the tendentiousness of the evaluation information is neutrality.
For example, first threshold is set for -0.5, second threshold 0.5, then, " I feels this commodity very not in upper example
It is good, simply too poor " emotional value of this evaluation information is -4, which is less than first threshold, then this evaluation information
Tendentiousness is commented for difference.
Step S207: whether the tendentiousness for judging the evaluation information of user's publication in each user community respectively is consistent
Favorable comment or it is consistent it is poor comment, if so, thening follow the steps S208, otherwise process terminates.
Step S208: the tendentiousness for determining the evaluation information of user's publication is consistent favorable comment or the consistent poor user community commented
There are group's brush single acts.
User community with commodity Assessment Rate higher (be higher than preset threshold) is often appeared in simultaneously and is commented different commodity
In valence, commodity are evaluated, if with the user community pair of commodity Assessment Rate higher (being higher than preset threshold) between some user
The evaluation information tendentiousness of shiploads of merchandise is identical, i.e., the member in user community be simultaneously to the evaluations of multiple commodity favorable comment or
Difference is commented, then judges that this user community exists and evaluate the single behavior of brush by commodity, then process terminates.
The embodiment of the present invention analyzes shiploads of merchandise evaluation information by intelligent algorithm, is found using GN algorithm
With the higher user community of commodity Assessment Rate, each user community is then analyzed to the evaluation information of commodity by emotional semantic,
It was found that evaluating the user community that consistent favorable comment or difference are commented to commodity, may determine that the user community based on above method, there are commodity
Group's brush single act in evaluation substantially increases without artificial report and judges whether it is the single efficiency of evaluation brush.
Fig. 3 is the main modular schematic diagram of the device of abnormal behaviour in judgement business activity according to an embodiment of the present invention.
As shown in figure 3, the device 300 of abnormal behaviour specifically includes that identification in the judgement business activity of the embodiment of the present invention
Module 301, analysis module 302, determination module 303.
Identification module 301, for belonging to the user of same user community according to the incidence relation identification between user.
Specifically, identification module 301 is used to determine the incidence relation between user according to community structure discovery algorithm, and deletes
With business subject evaluation rate be less than or equal to preset threshold two users between incidence relation, complete the deletion operation it
Afterwards, the user of same user community is belonged to according to the current incidence relation identification of user.
It wherein, is two users to two use of the sum of evaluation quantity of identical services object and this with business subject evaluation rate
Ratio of the family to all evaluation quantity summations of business object.
Analysis module 302, for analyzing the evaluation information of user's publication in each user community, to determine evaluation
The tendentiousness of information.
Specifically, analysis module 302 is used to carry out emotional semantic to the evaluation information of user's publication in each user community
It analyzes, and calculates the emotional value of every evaluation information according to the result that emotional semantic is analyzed, then determine every according to emotional value
The tendentiousness of evaluation information.
Every evaluation information includes one or more sense-groups.
Also, analysis module 302 specifically may include emotional semantic analysis module, emotional value computing module, tendentiousness determination
Module.
Wherein, emotional semantic analysis module is for issuing user each in each user community according to preset segmentation methods
Evaluation information segmented, and emotional semantic point is carried out to the participle of the evaluation information according to preset sentiment analysis dictionary
Analysis, to obtain the negative word in the participle of evaluation information, degree word, the corresponding weighted value of emotion word.
Emotional value computing module is used to calculate each meaning according to negative word, degree word, the corresponding weighted value of emotion word
The sense-group emotional value of group, and according to the sum of each sense-group emotional value in every evaluation information, obtain the emotion of every evaluation information
Value.
Tendentiousness determining module is used to for the emotional value of every evaluation information being compared with first threshold, second threshold,
And first threshold is less than second threshold, if the emotional value of an evaluation information is less than first threshold, it is determined that the evaluation information
Tendentiousness is commented for difference；If the emotional value of an evaluation information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is preferably
It comments；If the emotional value of an evaluation information is greater than first threshold and is less than second threshold, it is determined that the tendentiousness of the evaluation information
For neutrality.
Determination module 303, the tendentiousness for the evaluation information when user's publication each in a user community is consistent favorable comment
Or it is consistent it is poor comment, then determine that there are abnormal behaviours in business activity for the user community.
Fig. 4, which is shown, can determine the method for abnormal behaviour or judgement business in business activity using the embodiment of the present invention
The exemplary system architecture 400 of the device of abnormal behaviour in activity.
As shown in figure 4, system architecture 400 may include terminal device 401,402,403, network 404 and server 405.
Network 404 between terminal device 401,402,403 and server 405 to provide the medium of communication link.Network 404 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 401,402,403 and be interacted by network 404 with server 405, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 401,402,403
The application of page browsing device, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 401,402,403 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 405 can be to provide the server of various services, such as utilize terminal device 401,402,403 to user
The shopping class website browsed provides the back-stage management server supported.Back-stage management server can be to the evaluation information of user
Etc. data carry out the processing such as analyzing, and processing result is fed back into terminal device.
It should be noted that determining the method for abnormal behaviour in business activity generally by taking provided by the embodiment of the present invention
Business device 405 executes, and correspondingly, determines that the device of abnormal behaviour in business activity is generally positioned in server 405.
It should be understood that the number of terminal device, network and server in Fig. 4 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
Below with reference to Fig. 5, it illustrates the computer systems 500 for the server for being suitable for being used to realize the embodiment of the present application
Structural schematic diagram.Server shown in Fig. 5 is only an example, should not function and use scope band to the embodiment of the present application
Carry out any restrictions.
As shown in figure 5, computer system 500 includes central processing unit (CPU) 501, it can be read-only according to being stored in
Program in memory (ROM) 502 or be loaded into the program in random access storage device (RAM) 503 from storage section 508 and
Execute various movements appropriate and processing.In RAM 503, also it is stored with system 500 and operates required various programs and data.
CPU 501, ROM 502 and RAM 503 are connected with each other by bus 504.Input/output (I/O) interface 505 is also connected to always
Line 504.
I/O interface 505 is connected to lower component: the importation 506 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 508 including hard disk etc.；
And the communications portion 509 of the network interface card including LAN card, modem etc..Communications portion 509 via such as because
The network of spy's net executes communication process.Driver 510 is also connected to I/O interface 505 as needed.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 510, in order to read from thereon
Computer program be mounted into storage section 508 as needed.
Particularly, disclosed embodiment, the process described above with reference to flow chart may be implemented as counting according to the present invention
Calculation machine software program.For example, embodiment disclosed by the invention includes a kind of computer program product comprising be carried on computer
Computer program on readable medium, the computer program include the program code for method shown in execution flow chart.?
In such embodiment, which can be downloaded and installed from network by communications portion 509, and/or from can
Medium 511 is dismantled to be mounted.When the computer program is executed by central processing unit (CPU) 501, the system that executes the application
The above-mentioned function of middle restriction.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In this application, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In application, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in module involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part is realized.Described module also can be set in the processor, for example, can be described as: a kind of processor packet
Include identification module 301, analysis module 302, determination module 303.Wherein, the title of these modules is not constituted under certain conditions
Restriction to the module itself, for example, identification module 301 is also described as " for being known according to the incidence relation between user
Do not belong to the module of the user of same user community ".
As on the other hand, the present invention also provides a kind of computer-readable medium, which be can be
Included in equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes
Obtaining the equipment includes: to belong to the user of same user community according to the incidence relation identification between user；To in each user community
The evaluation information of user's publication is analyzed, with the tendentiousness of the determination evaluation information；It is respectively used according in the user community
Abnormal behaviour of the user community in the business activity described in the tendency sex determination of the evaluation information of family publication.It realizes and is not necessarily to people
To report, and substantially increasing that judging whether is the technical effect for evaluating the single efficiency of brush.
Technical solution according to an embodiment of the present invention belongs to same user community according to the incidence relation identification between user
User；The evaluation information of user's publication in each user community is analyzed, to determine the tendentiousness of evaluation information；When a use
In the group of family the tendentiousness of the evaluation information of each user publication be consistent favorable comment or it is consistent it is poor comment, then determine the user community in industry
There are abnormal behaviours in business activity.So that not needing to can be improved by artificially reporting and determine abnormal behaviour in business activity
Efficiency.
Above-mentioned specific embodiment, does not constitute a limitation on the scope of protection of the present invention.Those skilled in the art should be bright
It is white, design requirement and other factors are depended on, various modifications, combination, sub-portfolio and substitution can occur.It is any
Made modifications, equivalent substitutions and improvements etc. within the spirit and principles in the present invention, should be included in the scope of the present invention
Within.
Claims (14)
1. a kind of method for determining abnormal behaviour in business activity characterized by comprising
Belong to the user of same user community according to the incidence relation identification between user；
The evaluation information of user's publication in each user community is analyzed, with the tendentiousness of the determination evaluation information；
According to user community described in the tendency sex determination of the evaluation information of user each in user community publication in the business
Abnormal behaviour in activity.
2. the method according to claim 1, wherein belonging to same user according to the incidence relation identification between user
The step of user of group, comprising:
The incidence relation between the user is determined according to community structure discovery algorithm；
Delete the incidence relation being less than or equal between two users of preset threshold with business subject evaluation rate；
After the operation for completing the deletion, belong to same user community according to the current incidence relation identification of the user
User；
Wherein, the same business subject evaluation rate are as follows: two users are to the sum of evaluation quantity of identical services object and this two
Ratio of the user to all evaluation quantity summations of business object.
3. the method according to claim 1, wherein in each user community user publication evaluation information into
Row analysis, with the tendentious step of the determination evaluation information, comprising:
Emotional semantic analysis is carried out to the evaluation information of user's publication in each user community；
The emotional value of every evaluation information is calculated according to the result that the emotional semantic is analyzed；
The tendentiousness of every evaluation information is determined according to the emotional value.
4. according to the method described in claim 3, it is characterized in that, in each user community user publication evaluation information into
The step of market sense semantic analysis, comprising:
It is segmented according to the evaluation information that preset segmentation methods issue user each in each user community；
Emotional semantic analysis is carried out to the participle according to preset sentiment analysis dictionary, to obtain the negative in the participle
Word, degree word, the corresponding weighted value of emotion word.
5. according to the method described in claim 4, it is characterized in that, every evaluation information includes one or more sense-groups,
The step of calculating the emotional value of every evaluation information according to the result that the emotional semantic is analyzed, comprising:
The sense-group emotional value of each sense-group is calculated according to the negative word, degree word, the corresponding weighted value of emotion word；
According to the sum of each sense-group emotional value in every evaluation information, the emotion of every evaluation information is obtained
Value.
6. according to the method described in claim 3, it is characterized in that, determining every evaluation information according to the emotional value
Tendentious step, comprising:
The emotional value of every evaluation information is compared with first threshold, second threshold, the first threshold is less than institute
Second threshold is stated,
If the emotional value of an evaluation information is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；
If the emotional value of an evaluation information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is favorable comment；
If the emotional value of an evaluation information is greater than first threshold and is less than second threshold, it is determined that the evaluation information inclines
Tropism is neutrality.
7. a kind of device for determining abnormal behaviour in business activity characterized by comprising
Identification module, for belonging to the user of same user community according to the incidence relation identification between user；
Analysis module, for analyzing the evaluation information of user's publication in each user community, with the determination evaluation letter
The tendentiousness of breath；
Determination module, user group described in the tendency sex determination of the evaluation information for being issued according to user each in the user community
Abnormal behaviour of the body in the business activity.
8. device according to claim 7, which is characterized in that the identification module is also used to:
The incidence relation between the user is determined according to community structure discovery algorithm；
Delete the incidence relation being less than or equal between two users of preset threshold with business subject evaluation rate；
After the operation for completing the deletion, belong to same user community according to the current incidence relation identification of the user
User；
Wherein, the same business subject evaluation rate are as follows: two users are to the sum of evaluation quantity of identical services object and this two
Ratio of the user to all evaluation quantity summations of business object.
9. device according to claim 7, which is characterized in that the analysis module is also used to:
Emotional semantic analysis is carried out to the evaluation information of user's publication in each user community；
The emotional value of every evaluation information is calculated according to the result that the emotional semantic is analyzed；
The tendentiousness of every evaluation information is determined according to the emotional value.
10. device according to claim 9, which is characterized in that the analysis module includes emotional semantic analysis module, institute
Emotional semantic analysis module is stated to be used for:
It is segmented according to the evaluation information that preset segmentation methods issue user each in each user community；
Emotional semantic analysis is carried out to the participle according to preset sentiment analysis dictionary, to obtain the negative in the participle
Word, degree word, the corresponding weighted value of emotion word.
11. device according to claim 10, which is characterized in that every evaluation information includes one or more meanings
Group,
The analysis module further includes emotional value computing module, and the emotional value computing module is used for:
The sense-group emotional value of each sense-group is calculated according to the negative word, degree word, the corresponding weighted value of emotion word；
According to the sum of each sense-group emotional value in every evaluation information, the emotion of every evaluation information is obtained
Value.
12. device according to claim 9, which is characterized in that the analysis module includes tendentiousness determining module, described
Tendentiousness determining module is used for:
The emotional value of every evaluation information is compared with first threshold, second threshold, the first threshold is less than institute
Second threshold is stated,
If the emotional value of an evaluation information is less than first threshold, it is determined that the tendentiousness of the evaluation information is commented for difference；
If the emotional value of an evaluation information is greater than second threshold, it is determined that the tendentiousness of the evaluation information is favorable comment；
If the emotional value of an evaluation information is greater than first threshold and is less than second threshold, it is determined that the evaluation information inclines
Tropism is neutrality.
13. a kind of electronic equipment characterized by comprising
One or more processors；
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors
Realize such as method as claimed in any one of claims 1 to 6.
14. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
Such as method as claimed in any one of claims 1 to 6 is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711260817.2A CN109886702A (en) | 2017-12-04 | 2017-12-04 | The method and apparatus of abnormal behaviour in a kind of judgement business activity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711260817.2A CN109886702A (en) | 2017-12-04 | 2017-12-04 | The method and apparatus of abnormal behaviour in a kind of judgement business activity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109886702A true CN109886702A (en) | 2019-06-14 |
Family
ID=66923066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711260817.2A Pending CN109886702A (en) | 2017-12-04 | 2017-12-04 | The method and apparatus of abnormal behaviour in a kind of judgement business activity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886702A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427566A (en) * | 2019-07-23 | 2019-11-08 | 复旦大学 | A kind of service management system based on community system |
| CN112330373A (en) * | 2020-11-30 | 2021-02-05 | 中国银联股份有限公司 | User behavior analysis method and device and computer readable storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102004999A (en) * | 2010-12-06 | 2011-04-06 | 中国矿业大学 | Behaviour revenue model based collusion group identification method in electronic commerce network |
| CN103679462A (en) * | 2012-08-31 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Comment data processing method and device and searching method and system |
| CN104867017A (en) * | 2015-05-16 | 2015-08-26 | 成都数联铭品科技有限公司 | Electronic commerce client false evaluation identification system |
| CN104866468A (en) * | 2015-04-08 | 2015-08-26 | 清华大学深圳研究生院 | Method for identifying false Chinese customer reviews |
| CN105809451A (en) * | 2016-02-29 | 2016-07-27 | 江苏大学 | Big data based e-commerce company evaluating, analyzing and predicting method and system for online shopping |
| CN106022878A (en) * | 2016-05-19 | 2016-10-12 | 华南理工大学 | Community comment emotion tendency analysis-based mobile phone game ranking list construction method |
| CN106059847A (en) * | 2016-04-08 | 2016-10-26 | 久远谦长（北京）技术服务有限公司 | User ordering behavior detection method and device |
| CN106127505A (en) * | 2016-06-14 | 2016-11-16 | 北京众成汇通信息技术有限公司 | The single recognition methods of a kind of brush and device |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
| CN107305665A (en) * | 2016-04-19 | 2017-10-31 | 李辉 | It is a kind of to differentiate wash sale, prevent the single method and device of brush |
-
2017
- 2017-12-04 CN CN201711260817.2A patent/CN109886702A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102004999A (en) * | 2010-12-06 | 2011-04-06 | 中国矿业大学 | Behaviour revenue model based collusion group identification method in electronic commerce network |
| CN103679462A (en) * | 2012-08-31 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Comment data processing method and device and searching method and system |
| CN104866468A (en) * | 2015-04-08 | 2015-08-26 | 清华大学深圳研究生院 | Method for identifying false Chinese customer reviews |
| CN104867017A (en) * | 2015-05-16 | 2015-08-26 | 成都数联铭品科技有限公司 | Electronic commerce client false evaluation identification system |
| CN105809451A (en) * | 2016-02-29 | 2016-07-27 | 江苏大学 | Big data based e-commerce company evaluating, analyzing and predicting method and system for online shopping |
| CN106059847A (en) * | 2016-04-08 | 2016-10-26 | 久远谦长（北京）技术服务有限公司 | User ordering behavior detection method and device |
| CN107305665A (en) * | 2016-04-19 | 2017-10-31 | 李辉 | It is a kind of to differentiate wash sale, prevent the single method and device of brush |
| CN106022878A (en) * | 2016-05-19 | 2016-10-12 | 华南理工大学 | Community comment emotion tendency analysis-based mobile phone game ranking list construction method |
| CN106127505A (en) * | 2016-06-14 | 2016-11-16 | 北京众成汇通信息技术有限公司 | The single recognition methods of a kind of brush and device |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
Non-Patent Citations (1)
| Title |
|---|
| 涂海丽等: "基于在线评论的游客情感分析模型构建", 《现代情报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427566A (en) * | 2019-07-23 | 2019-11-08 | 复旦大学 | A kind of service management system based on community system |
| CN112330373A (en) * | 2020-11-30 | 2021-02-05 | 中国银联股份有限公司 | User behavior analysis method and device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106886518B (en) | Microblog account number classification method | |
| CN109697641A (en) | The method and apparatus for calculating commodity similarity | |
| CN109636430A (en) | Object identifying method and its system | |
| CN107944911A (en) | A kind of recommendation method of the commending system based on text analyzing | |
| CN110647522A (en) | Data mining method, device and system | |
| CN109325121A (en) | Method and apparatus for determining the keyword of text | |
| CN110276065A (en) | A kind of method and apparatus handling goods review | |
| CN110111167A (en) | A kind of method and apparatus of determining recommended | |
| CN110119445A (en) | The method and apparatus for generating feature vector and text classification being carried out based on feature vector | |
| CN112733042A (en) | Recommendation information generation method, related device and computer program product | |
| CN110362815A (en) | Text vector generation method and device | |
| CN108595448A (en) | Information-pushing method and device | |
| CN110309293A (en) | Text recommended method and device | |
| CN113326420A (en) | Question retrieval method, device, electronic equipment and medium | |
| CN111861596A (en) | Text classification method and device | |
| CN109886702A (en) | The method and apparatus of abnormal behaviour in a kind of judgement business activity | |
| CN107291835A (en) | A kind of recommendation method and apparatus of search term | |
| CN110516033A (en) | A kind of method and apparatus calculating user preference | |
| CN109754273A (en) | The method and apparatus for promoting any active ues quantity | |
| CN112667770A (en) | Method and device for classifying articles | |
| Blenn et al. | Context-sensitive sentiment classification of short colloquial text | |
| CN112418258A (en) | Feature discretization method and device | |
| CN110766431A (en) | Method and device for judging whether user is sensitive to coupon | |
| US10733221B2 (en) | Scalable mining of trending insights from text | |
| CN110020131A (en) | A kind of method and apparatus arranging commodity |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |