CN108494669A - A kind of method and system from social object to user that recommending - Google Patents
A kind of method and system from social object to user that recommending Download PDFInfo
- Publication number
- CN108494669A CN108494669A CN201810295820.6A CN201810295820A CN108494669A CN 108494669 A CN108494669 A CN 108494669A CN 201810295820 A CN201810295820 A CN 201810295820A CN 108494669 A CN108494669 A CN 108494669A
- Authority
- CN
- China
- Prior art keywords
- communication
- user
- social
- time
- real
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/52—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail for supporting social networking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]
- H04L51/046—Interoperability with other network applications or services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of methods from social object to user that recommending, including：Obtain the expression information of user；According to the communication score information in the historical communication record storehouse of the expression information and the user, the social object of recommendation is generated；The social object of the recommendation is presented to the user, and real-time social communication is established with the social object of the recommendation.In addition, the invention also discloses a kind of systems from social object to user that recommending.The problem that the way of recommendation that can efficiently solve existing social recommendation system through the invention is not smart enough, accuracy is not high accurately recommends social object to user to meet active user's psychological needs.
Description
Technical field
The present invention relates to social communications to recommend field, more particularly to a kind of to recommend the method for social object to user and be
System.
Background technology
The rhythm of the modern life is getting faster, and operating pressure is increasing, and people will produce the idle heart for social activity
Reason, the people for being detached from social circle are more prone to the negative emotions such as anxiety, and the mental health and social interaction for people generate
Prodigious negative effect.
Popularizing for instant communication software, facilitates the social communication of people.Conventional social recommendation system is based on user's
Address list and message registration, additionally provide periodically prompting or the birthday in red-letter day reminds.
But the existing way of recommendation can only be recommended for festivals or holidays or the specific date (birthday etc.) of setting, no
Enough intelligence can not be directed to user demand and carry out accurate recommendation；The way of recommendation is fixed, and the object recommended, which is difficult to meet user, works as
Preceding psychological needs.
To sum up, there are not smart enough, inaccurate defect and limitations for existing social communication system.
Invention content
The present invention provides a kind of method and system from social object to user that recommending, to solve existing social recommendation system
The way of recommendation is not smart enough, technical problem that accuracy is not high.It is specific that technical scheme is as follows：
On the one hand, the invention discloses a kind of methods from social object to user that recommending, including：
S100 obtains the expression information of user；
S200 is pushed away according to the communication score information in the historical communication record storehouse of the expression information and the user, generation
The social object recommended；
The social object of the recommendation is presented to the user by S300, and establishes real-time society with the social object of the recommendation
Traffic is believed.
Said program is commented using the current psychology of the current expression information reflection user of user according in historical communication record storehouse
Divide information, be intelligently directed to the scheme that user demand carries out accurate recommendation social activity object, efficiently solves existing social logical
There are not smart enough, inaccurate defect and limitations for letter system.
Preferably, the historical communication record storehouse also includes the time of each historical communication；The step S200 is according to institute
The communication score information in the historical communication record storehouse of expression information and the user is stated, the social object for generating recommendation specifically wraps
It includes：S210 is when the expression information is happy or normal, according to the communication score information to the historical communication record storehouse
In communications records be ranked up, the current farthest communication object of the forward and historical communication time gap of sorting is as recommendation society
Hand over object.
Preferably, the historical communication record storehouse also includes the initial mood of each historical communication；The step S200 roots
According to the communication score information in the historical communication record storehouse of the expression information and the user, the social object tool of recommendation is generated
Body includes：S220 when the expression information be it is sad or angry when, searched in the historical communication record storehouse initial mood with
The expression information is identical and the highest communications records of scoring, using the corresponding communication object of the communications records as recommendation society
Hand over object.
In above-mentioned technical proposal, the user mood is judged according to the expression information, will be used not according to user mood
Same suggested design.When the user mood is happy or normal, using the recommendation based on help user maintenance social networks
Scheme；When the user mood is sad or angry, the suggested design based on unhealthy emotion is mediated using user is helped.In this way
Suggested design can solve existing social intercourse system not intelligence, inaccurate defect and limitation.
Preferably, further include after the step S300：
S400 carries out real-time state monitoring to the real-time social communication；
The communication that S500 calculates this real-time social communication according to the data monitored in real time is scored；
The communication scoring update of this real-time social communication record and the social communication is arrived the historical communication by S600
Record storehouse.
Score information in historical communication record storehouse is to recommend one of the foundation of social object, needs to constantly update.Using
Monitoring in real time is carried out to the real-time social communication and updates the scoring letter in historical communication record storehouse according to monitoring result
The function of accurately recommending social object according to user's current mood just may be implemented in the scheme of breath.
Preferably, the data monitored in real time include：Communication start when mood value, sign off when mood value,
Voice duration in call total duration, the mood value of user's t moment, user's communication；The step S500 is according to the real-time prison
The scoring that the data of survey calculate this real-time social communication specifically includes：
S510 calculates the user this real-time social communication according to the data monitored in real time, by following equation
Communication scoring：
Wherein, S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is
Time variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；m
(t) be user described in t moment in communication process mood value；P is the percentage number shared by the voice of user in this communication
Value；K1, K2 are the proportionality coefficient constant of positive value.
The communication scoring of the above-mentioned scoring formula for establishing the real-time social communication of single, the real-time social communication of single includes that basis is closed
Set occurrence (B), social quality evaluation value (K1* (M2-M1) * T*P) and user emotion fluctuating value
Basis divides constant, effect to be that score value difference is excessively greatly different between preventing record, avoids system excessively sensitive.User emotion fluctuating value
For recording user's emotional change in communication process.Social quality evaluation value accumulates emotional change of the user in call
Divide and calculate, mood value influence to score of the time at the end of is bigger, can weigh the direction of mood transformation, mood is just
It can become larger to transformation value, negative sense transformation, value can become smaller.
On the other hand, the invention also discloses a kind of systems from social object to user that recommending, including：Expression obtains mould
Block, the expression information for obtaining user；Historical communication record storehouse, the historical communication for storing the user records, described
Historical communication record includes communication score information；Social analysis module, the expression for being obtained according to the expression acquisition module
Communication score information in information and the historical communication record storehouse of the user generates the social object of recommendation；Social recommendation mould
Block, the social object for the recommendation that the social analysis module generates to be presented to the user, and with the social activity of the recommendation
Object establishes real-time social communication.
Preferably, the historical communication record storehouse also records the time for having each historical communication；The social activity analysis module
It specifically includes：Expression analysis submodule, the expression information for analyzing the user that the expression acquisition module obtains；Sequence
Submodule, when the expression information for the user described in the Expression analysis submodule block analysis is happy or normal, according to described
Score information is ranked up the communications records in the historical communication record storehouse；Object generates submodule, is used for the row
Sequence submodule block sequencing is forward and the current farthest user of historical communication time gap is as the social object of recommendation.
Preferably, the historical communication record storehouse also records the initial mood for having each historical communication；The social analysis
Module further includes：Search submodule, the expression information for the user described in the Expression analysis submodule block analysis be it is sad or
When angry, search that initial mood is identical as the expression information and the highest communication of scoring in the historical communication record storehouse
Record, using the corresponding communication object of the communications records as the social object of recommendation.
Preferably, the system to the social object of user's recommendation further includes：Communication monitoring module, for described real-time
Social communication carries out real-time state monitoring；Communicate grading module, the data for being monitored in real time according to the communication monitoring module
Calculate the communication scoring of this real-time social communication；Module is updated storage, is used to record this real-time social communication and described
In the communication scoring update to the historical communication record storehouse of social communication.
Preferably, the data that the communication monitoring module monitors in real time include：Initial mood value, communication when communication starts
At the end of mood value, call total duration, the mood value of user's t moment, user communication in voice duration；The communication scoring
Module is specifically included according to the scoring that the data monitored in real time calculate this real-time social communication：The communication grading module
According to the data that the communication monitoring module monitors in real time, the user this real-time social communication is calculated by following equation
Communication scoring：
Wherein：S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is
Time variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；m
(t) be user described in t moment in communication process mood value；P is the percentage number shared by the voice of user in this communication
Value；K1, K2 are the proportionality coefficient constant of positive value.
The present invention is by using by the score information knot in the historical communication record storehouse of the current expression information of user and user
It closes, generates the social object for being suitble to user's current mood, the social object of recommendation is presented to user, and establish social in real time logical
Letter.The current psychology of the current expression information reflection user of user is intelligently directed to according to score information in historical communication record storehouse
User demand carries out accurate recommendation social activity object.Efficiently solving existing social communication system, there are not smart enough, not smart
Its limitation of true defect.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of flow chart of the embodiment of method to the social object of user's recommendation of the present invention；
Fig. 2 is a kind of flow chart of another embodiment of method to the social object of user's recommendation of the present invention；
Fig. 3 is a kind of flow chart of another embodiment of method to the social object of user's recommendation of the present invention；
Fig. 4 is a kind of block diagram of system embodiment to the social object of user's recommendation of the present invention；
Fig. 5 is a kind of block diagram of another embodiment of system to the social object of user's recommendation of the present invention；
Fig. 6 is a kind of block diagram of another embodiment of system to the social object of user's recommendation of the present invention；
Fig. 7 is a kind of flow chart of another embodiment of method to the social object of user's recommendation of the present invention.
Reference numeral：
10-- expression acquisition modules；20-- historical communication record storehouses；30-- social activity analysis modules；31-- Expression analysis submodules
Block；32-- sorting sub-modules；33-- objects generate submodule；34-- searches submodule；40-- social recommendation modules；50-- is communicated
Grading module；60-- updates storage module；70-- communication monitoring modules.
Specific implementation mode
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation
Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts
All other embodiment, shall fall within the protection scope of the present invention.
The present invention provides a kind of method from social object to user that recommending, and embodiment is as shown in Figure 1, include：
S100 obtains the expression information of user；
S200 is pushed away according to the communication score information in the historical communication record storehouse of the expression information and the user, generation
The social object recommended；
The social object of the recommendation is presented to the user by S300, and establishes real-time society with the social object of the recommendation
Traffic is believed.
In above-described embodiment, the expression information of user is got, and then can be used according to the current expression information reflection of user
The current psychology in family is intelligently directed to user demand and carries out accurately further according to the communication score information in historical communication record storehouse
Recommend social object, efficiently solving existing social communication system, there are not smart enough, inaccurate defect and limitations.
Another embodiment of the method for the present invention, on the basis of above-described embodiment, the historical communication record storehouse also includes every
The time of secondary historical communication, each historical communication initial mood etc..Specifically, the present embodiment is believed according to the different expressions of user
Breath takes different strategies to carry out social recommendation, realizes and precisely recommends.Specifically, as shown in Fig. 2, including：
S100 obtains the expression information of user；
S210 remembers the historical communication according to the communication score information when the expression information is happy or normal
Communications records in record library are ranked up, and the current farthest communication object of the forward and historical communication time gap of sorting is as pushing away
Recommend social object；
S220 when the expression information be it is sad or angry when, searched in the historical communication record storehouse initial mood with
The expression information is identical and the highest communications records of scoring, using the corresponding communication object of the communications records as recommendation society
Hand over object；
The social object of the recommendation is presented to the user by S300, and establishes real-time society with the social object of the recommendation
Traffic is believed.
The expression of user has reacted the heart condition of user at that time to a certain extent, therefore, by capturing user on the face
Expression can go out the mood of user with preliminary analysis.Existing various technical solutions may be used for the identification of expression.
The state of several characteristic portions such as eye, eyebrow, mouth and the face of user can be specifically captured, then passes through the calculating such as algorithm
Analyze the corresponding mood of the facial expression.For example, user's face expression is smiled, then the information as obtained by the expression is to use
Family is currently relatively more happy；One's eyebrows knit in a frown and mouth sagging for user's face, then the corresponding information of the expression is that mood is sad.So
Afterwards, when judge user's expression information for it is happy or normal when, according to the communication score information in historical communication record storehouse to history
Communications records in communications records library are ranked up, by the communication that communication scoring is forward and historical communication time gap is currently farthest
Object is as the social object of recommendation.Or when judge user mood for it is sad or angry when, first looked for from historical communication record storehouse
Go out the initial mood of the communication beginning historical communication record consistent with the current expression information of user, that is to say, that first find out communication note
Initial mood is also sad or angry communications records in record, then by these communications records found out according to communication scoring
Height is ranked up, and the corresponding social object of the highest communications records that will finally score is as the social object of recommendation.To help
User mediates unhappy in the heart.
Certainly, we can also recommend multiple objects to be selected for user, such as in user when recommending social object
When sad or angry, mood is started according to communication and the communications records in historical communication record storehouse are arranged in communication scoring
Sequence, mood is consistent with expression information before preset value is communicated and communicate the forward communication object of scoring right as recommendation social activity
As.For example, ought be gone out according to user's face Expression analysis, user is sad at this time, then being searched in the historical communication record storehouse of user
The initial mood that communication starts is sad sad communications records, if historical communication record one co-exists in 100 records, is then looked into
It is that sad sad record amounts to 50 to look for the initial mood of user；So, we are just by find out 50 records according to communication
The height of scoring is ranked up, then will score again it is high before 3 bit walk objects as social recommendation object recommendation to user, so
The recommended again with user's selection establishes real-time Communication for Power afterwards.For another example, when user is happy, social object is recommended to use help
Suggested design based on user maintenance social networks is first ranked up log according to communication scoring, then again will sequence
Preceding preset value record is forward by TOP SCORES and historical communication time gap is current farthest according to historical communication time-sequencing
Communication object, which is used as, recommends social object, specifically, preceding 20 communications records for such as selecting marking and queuing forward first, then
Corresponding communication object is found according to this 20 communications records, then according still further to the logical of these communication objects and the user
The distance of letter time is ranked up, that is to say, that time that distance and the user last time communicate longer communication object is more arranged
In front, after according to this rule compositor, then taking in the result of last sequence front several to recommend user (such as will sequence
3 communication objects the 1st, 2,3 recommend user).The communication object screened in this way, on the one hand, scoring is to compare to lean on
Preceding, illustrate that the relationship of these communication objects and user should be relatively good, but they are for a long time without communication, if long
If time does not contact, it may slowly desalinate this social networks, therefore, in order to help these good social activities of user maintenance to close
System, then can recommend these scorings higher to user but communication object that is not contacting for a long time, to play side user maintenance
The effect of good social networks.
Certainly, add up with the time, the communications records in the communications records library of user can be more and more, communication scoring
It accordingly can be more and more.And over time, perhaps user also can constantly have new friend, at this moment if by history
If communications records in communications records library all carry out marking and queuing, the possible workload of one side and time-consuming president, another party
Face is also not necessarily the optimal recommended of present user.Therefore, we can be carried out preliminary by setting time or quantity
Screening.For example, only select nearly one month or nearly bimestrial communications records in user's history communications records libraries, then further according to
The expression information of user, which carries out corresponding communication marking and queuing, to be recommended.
In above-mentioned technical proposal, the mood of the user can be tentatively judged according to the expression information, according to user's table
The difference (being equivalent to mood difference) of feelings will use different suggested designs.When user's expression information is happy or normal
When, the earlier social object that do not contact for a long time but score can be recommended, to help user maintenance social networks；When described
When user's expression information is sad or angry, the suggested design based on unhealthy emotion is mediated using user is helped.Such recommendation
Scheme can solve existing social intercourse system not intelligence, inaccurate defect and limitation.
Another embodiment of the method for the present invention calculates each communication using scoring according to the communication in historical communication record storehouse
Object TOP SCORES once leads to recommend the technical solution of social object according to TOP SCORES due to communicating every time
Letter scoring, and the same social object may have multiple communication with user, that is to say, that it has repeatedly communication scoring and is stored in
In historical communication record storehouse, therefore, can before each recommend, in optional preset time (such as nearly half a year, nearly three months or
Nearly one month of person etc.) communications records, scored according to all communications of the social object in this section of preset time to calculate this
The TOP SCORES of social object；Recommendation is ranked up to the TOP SCORES again below.Likewise, can be believed according to the expression of user
Breath, carrys out intelligent recommendation in conjunction with TOP SCORES information etc..For example, when the expression information of user is happy or normal, select total
Body scores the current farthest communication object of forward and historical communication time gap as the social object of recommendation；When the expression of user is believed
When breath is sad or angry, first according to the record of the initial mood value communicated every time, being searched in the historical communication record storehouse
All initial mood values communications records consistent with user's current mood value in the preset period；Then according to lookup
The communications records arrived obtain corresponding communication object；These communication objects got are carried out according to respective TOP SCORES again
Sequence, then will score it is highest recommend user, to help user discongest in the minds of it is depressed.
And for the calculating of TOP SCORES, then various ways may be used, such as the several ways that we enumerate below：
(1) average value of all communications scoring of each communication object in preset period is calculated as the total of the communication object
Body scores.
(2) according to the communications records in historical communication record storehouse in preset period, corresponding communication object, then root are found
Different weights are distributed according to the duration of each communication distance current time of communication object, finally seek the communication object pre-
If all communication scorings and the product of corresponding weights and as the communication object TOP SCORES in period.
(3) communications records in historical communication record storehouse in preset period are selected, corresponding communication object is found, will lead to
Believe TOP SCORES of the recent communication scoring of object distance current time as the communication object.
We are by taking preset period is nearly one month as an example, we obtain nearly one from historical communication record storehouse first
Then communications records in month obtain corresponding communication object according to these communications records, for example the communication object obtained has
Three, Li Si, king five etc..By taking the TOP SCORES for calculating Zhang San as an example, for example Zhang San had in this nearly two months with the user
It communicates three times, is respectively：At 2018 2 months 2 days 14, corresponding communication scoring is 85 points；It is corresponding when 15 days 11 March in 2018
Communication scoring be 80 points；When 23 days 21 March in 2018, corresponding communication scoring is 90 points.We start to calculate Zhang San now
TOP SCORES, if the scheme of employing mode (1), can selecting Zhang San, this communicates the average value of scoring as Zhang San's three times
TOP SCORES, that is to say, that the TOP SCORES of Zhang San is (85+80+90)/3=85 points.If the scheme of employing mode (2)
Words, first need, according to different weights are distributed apart from current call duration time, then to calculate again, longer apart from current time
Communications records distribution weights it is also lower, the current closer communications records of distance, then the weights distributed are then higher；Such as it is logical
The TOP SCORES of Zhang San is after crossing distribution weights：85*0.2+80*0.3+90*0.5=86 points.If the scheme of employing mode (3)
If, then it can directly select TOP SCORES of the communication scoring of recent communications records as Zhang San, that is to say, that
Three TOP SCORES is 90 points.No matter which kind of mode used, as long as the communication object of this period all comes in the same way
Calculate TOP SCORES, then can according to the height of TOP SCORES come to each communication object into being ranked, finally in conjunction with
The expression information of user is recommended using corresponding Generalization bounds.
Said program is not to be recommended to be ranked up according to the height of certain communication scoring merely, but combine logical
The multiple communication of letter object is scored to carry out comprehensive consideration, to, TOP SCORES more accurate to the analysis of the communication object
Also the communication object and the cohesion of user can be more embodied, to but also last recommendation results are also more in line with user's
Psychology, recommendation is more intelligent, improves the satisfaction of user.
Another embodiment of the present invention, based on any of the above embodiments, as shown in figure 3, including：
S100 obtains the expression information of user；
S200 is pushed away according to the communication score information in the historical communication record storehouse of the expression information and the user, generation
The social object recommended；
The social object of the recommendation is presented to the user by S300, and establishes real-time society with the social object of the recommendation
Traffic is believed；
S400 carries out real-time state monitoring to the real-time social communication；
The communication that S500 calculates this real-time social communication according to the data monitored in real time is scored；
The communication scoring update of this real-time social communication record and the social communication is arrived the historical communication by S600
Record storehouse.
Communication score information in historical communication record storehouse is to recommend one of the foundation of social object, needs to constantly update.
Using to the real-time social communication monitor in real time and according to institute's commentary in monitoring result update historical communication record storehouse
The function of accurately recommending social object according to user's current mood just may be implemented in the scheme for dividing information.
In above-described embodiment, the data monitored in real time include：Communication start when mood value, sign off when the heart
Voice duration etc. in feelings value, call total duration, the mood value of user's t moment, user's communication.Specifically, above-mentioned steps S500
The scoring for calculating this real-time social communication according to the data monitored in real time specifically includes：S510 is according to the real-time monitoring
Data, the communication that the user this real-time social communication is calculated by following equation scores：
Wherein, S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is
Time variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；m
(t) be user described in t moment in communication process mood value；P is the percentage number shared by the voice of user in this communication
Value；K1, K2 are the proportionality coefficient constant of positive value.
In above-mentioned formula, the mood value of user is obtained according to the expression of user, by the facial table for analyzing user
Feelings information can obtain the mood of user at that time.For example user mood can be divided into 4 classes and assign definition numerical value by we：Just
Normal 1, happy 2, sad -1, indignation -2；These numerical value are mood values.
Above-mentioned formula is mainly made of three parts：A part is that constant, that is, B are divided in basis；Second part is K1* (M2-
M1) * T*P, K1 are positive proportionality coefficients, and user emotion changes (M2-M1) when being positive, this part is positive value, and emotional change is negative
Xiang Shi, this partial value are negative value.That is if user is to turn to positive value from mood negative value, increase scoring, conversely, can then reduce
Scoring.Part III isK2 is positive proportionality coefficient, by mood of the user in call
Variation carries out integral and calculating, and mood value influence to score of the time at the end of is bigger, can weigh the side of mood transformation
To mood positive transition value can become larger, and negative sense transformation, value can become smaller.For example, user and Zhang San newly establish communication, the time started is
2018-02-0615：45, end time 2018-02-0616：03, according to monitored results see user communication before mood be it is happy,
User mood is indignation in communication, and user mood is back to normal after sign off.User speech percentage is 30%.Then
The communication of the secondary communication is scored：Wherein
K1, K2 are the proportionality coefficient constant of positive value, if it is 67 points to be computed rear S, i.e. the communication of user and the current real-time Communication for Power of Zhang San
Scoring is 67 points.
On the other hand, the invention also discloses a kind of systems from social object to user that recommending, specifically, as shown in figure 4,
Including：Expression acquisition module 10, the expression information for obtaining user；Historical communication record storehouse 20, for storing the user
Historical communication record, historical communication record includes communication score information；Social analysis module 30, for according to the table
Communication score information in the historical communication record storehouse 20 of expression information and the user that feelings acquisition module 10 obtains, generation push away
The social object recommended；Social recommendation module, the society for the recommendation that the social analysis module 30 generates to be presented to the user
Object is handed over, and real-time social communication is established with the social object of the recommendation.
Specifically, first, the expression information of user, mobile phone, tablet computer or other electricity are obtained by expression acquisition module
Sub- smart machine can obtain the expression information of user, and be sent to social analysis module；Social analysis module is believed according to expression
Communication score information in the historical communication record storehouse of breath and user, generates the social object of recommendation；Social recommendation module to
The social object recommended is presented in family, and establishes real-time social communication with the social object recommended；Historical communication record storehouse is for remembering
The historical communication record at family is employed, historical communication record just contains communication score information, that is to say, that can be to communicating every time
It all gives a mark, then exists in historical communication record storehouse, convenient for being referred to as recommendation when recommending social object using rear line.
Another embodiment of present system, as shown in figure 5, on the basis of above system embodiment, the history is logical
Letter record storehouse 20 also records the initial mood etc. of the time, each historical communication that have each historical communication；The social analysis mould
Block 30 specifically includes：Expression analysis submodule 31, the expression for analyzing the user that the expression acquisition module 10 obtains
Information；Sorting sub-module 32, for analyzing the expression information of the user when the Expression analysis submodule 31 as happily or just
Chang Shi is ranked up the communications records in the historical communication record storehouse 20 according to institute's scoring information；Object generates submodule
Block 33, for the sorting sub-module 32 to be sorted the current farthest user of forward and historical communication time gap as recommendation society
Hand over object.
The social activity analysis module 30 further includes：Submodule 34 is searched, for being analyzed when the Expression analysis submodule 31
When the expression information of the user is sad or angry, initial mood and the table are searched in the historical communication record storehouse 20
Feelings information is identical and the highest communications records of scoring, using the corresponding communication object of the communications records as recommending social activity object.
Social analysis module generates two kinds of schemes for recommending social object according to user's expression information, when user mood is to open
The heart or it is normal when, using help user maintenance social networks based on suggested design；When user mood is sad or angry, adopt
The suggested design based on unhealthy emotion is mediated with user is helped.For example, user's expression is to be raised under the corners of the mouth, Expression analysis submodule is sentenced
Being completely cured, it is currently sader sad to get user according to the expression, and it is sad that initial mood is searched in historical communication record storehouse
Communications records, then will find initial mood is that sad communications records are ranked up according to the height of communication scoring, by it
The middle highest communication object of scoring recommends user.
Preferably, in any of the above-described embodiment, the system to the social object of user's recommendation further includes：Communication monitoring mould
Block 70, for carrying out real-time state monitoring to the real-time social communication；Grading module 50 is communicated, for being supervised according to the communication
Survey the communication scoring that the data that module 70 monitors in real time calculate this real-time social communication；Module 60 is updated storage, is used to incite somebody to action this
In the communication scoring update to the historical communication record storehouse 20 of secondary real-time social communication record and the social communication.
Score information in historical communication record storehouse is to recommend one of the foundation of social object, needs to constantly update.Communication
Monitoring modular using real-time social communication is monitored in real time, and communication grading module according to monitoring as a result, calculating this in real time
The scoring of secondary real-time social communication updates storage module according to the score information in monitoring result update historical communication record storehouse
The function of accurately recommending social object according to user's current mood just may be implemented in scheme in this way.
Preferably, the data that the communication monitoring module monitors in real time include：Initial mood value, communication when communication starts
At the end of mood value, call total duration, the mood value of user's t moment, user communication in voice duration；The communication scoring
Module is specifically included according to the scoring that the data monitored in real time calculate this real-time social communication：The communication grading module
According to the data that the communication monitoring module monitors in real time, the user this real-time social communication is calculated by following equation
Communication scoring：
Wherein：S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is
Time variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；m
(t) be user described in t moment in communication process mood value；P is the percentage number shared by the voice of user in this communication
Value；K1, K2 are the proportionality coefficient constant of positive value.
The last one integrated embodiment of the present invention provides a kind of system from social object to user that recommending, the system
The recommendation method of the present invention can be used.The social recommendation system is a software systems, needs to be integrated in mobile phone, tablet computer
Or use that (these equipment must include camera, and microphone, communication apparatus meets performance requirement in other smart machines
CPU and enough memory spaces).
Specifically, as shown in fig. 6, the system includes historical record library 1, social address list 2, communication module 3, vision module
4, sense of hearing module 5, center module 6.Wherein：
Historical record library 1 records the log of each user, including the mood before communication object, communication duration, communication
The mood states after mood states, communication, user speech percentage and TOP SCORES in state, communication etc..
Social address list 2, user need the social object information of oneself being input in social address list, and content includes surname
Name, the information such as gender can be the address list in mobile phone, can also include intimate degree optionally.
Communication module 3 completes video calling and voice call function, mobile phone or the communication of other intelligent terminals can also be used
Function.
Vision module 4 is equivalent to the expression shape change of monitoring user, analyzes user's for identifying user identity and expression
Psychology.
Sense of hearing module 5, for identifying in speech intonation and communication process in user's communication process shared by the voice of user
The percentage of entire communication time.That is statistics is monitored to the voice in communication process.
Center module 6, mainly there is following functions：
1) in communication process, the intonation variation, voice variation and voice that record user in real time by sense of hearing module account for pair
The variation of the percentage of words process, and according to user's expression information of vision module monitoring, obtain corresponding mood value etc.；
After sign off, by above-mentioned real-time record monitoring information and total communication duration, the evaluation of this communication is provided, will be led to
Storage in " historical record library " is put into news record and evaluation.
2) define the time of social recommendation, for example, be set by the user a period for allowing suggestion (such as on every Saturdays under
2 points to 6 points of noon), when the face of system detectio to user, start to generate social suggestion, and notify user.
3) right according to user's current mood (passing through vision module Expression Recognition) when needing to generate social suggestion object
The log and evaluation stored in " historical record library " is analyzed, and generates social suggestion object to user.
The flow that social suggesting system for wearing carries out social object recommendation is introduced below in conjunction with above-mentioned commending system, as shown in fig. 7,
Include the following steps：
Step S0, software brings into operation.
Step S1, user's registration, typing social activity address list information, face information, voiceprint are carried out, and setting allows to build
The time interval of view.
Whether step S2, check current time, in the period allowed, is：Step S3 is executed, it is no：Execute step S2.
Step S3, it checks that this allows the period not suggest to user, is：Step S4 is executed, it is no：Execute step
S2。
Step S4, it checks and recognizes user's face information, be：Step S5 is executed, it is no：Execute step S2.
Step S5, the current expression of user is obtained
Step S6, retrieval user " social address list " and " historical record library " reads address list information and historical record.
Step S7, it is recorded according to user's current mood and historical communication, provides the social object suggestion of recommendation.
Step S8, it checks whether user receives suggestion, is：Step S9 is executed, it is no：Execute step S2.
Step S9, video speech communication is proceeded by.
Step S10, start to monitor user's expression shape change, the variation of user's intonation.
Step S11, voice percentage of the record user in dialogue.
Step S12, after communication, pass through the expression shape change of user, the monitoring of tone variation and voice percentage
Record, scores to communication quality, and generate record and be stored in " historical record library ".
Step S7 is related to the scoring of historical communication record to step S12, is needed in step 7 using these scorings, raw
Suggest at social activity, can score each communication quality, be stored in " historical record library ", step S7 is to make in step S12
User, step S12 are the producers.
Here, illustrating the code of points of step S12 first.
It is directed to described in step S12, a kind of code of points is as follows, and user mood is divided into 4 classes and assigns definition number
Value：Normal 1, happy 2, sad -1, indignation -2；If total air time is T, t is time variable, heart when M2 is sign off
Feelings value, M1 are mood values when communication starts, and user mood value when m (t) times are t, P is shared by the voice in user's communication
Percentages, K1, K2 are the proportionality coefficient constant of positive value, and S is communication score, divides constant based on B.The calculating of total score S
Formula has：
Formula Parsing：
S points form for three parts score value：
First part：It is that score value difference is excessively greatly different between preventing record to divide constant, effect based on B, avoids system
It is excessively sensitive.
Second part：K1* (M2-M1) * T*P, K1 are positive proportionality coefficients, and user emotion changes (M2-M1) when being positive,
This part is positive value, and when emotional change is negative sense, this partial value is negative value；When positive change, T and P are bigger, and score value is bigger；
When negative sense changes, T and P are bigger, and score value is smaller.
Part III：K2 is positive proportionality coefficient, by feelings of the user in call
Thread variation carries out integral and calculating, and mood value influence to score of the time at the end of is bigger, can weigh mood transformation
Direction, mood positive transition value can become larger, and negative sense transformation, value can become smaller.
Three parts score value is taken and the scoring specifically communicated.
Illustrate step S7 how using these scorings now：
It is directed to described in step S7, the rule of social recommendation object is generated according to user's current mood and historical communication record
It is then as follows：
When the mood of user is happy and normal, marking and queuing is carried out during historical communication is recorded, is found out preceding
In being recorded in 50%, the object being not in contact at most recommends user, for helping user maintenance social networks.
When the mood of user is sad and angry, the note that initial mood is current mood is found in historical communication record
Record finds the highest object that scores in these records, recommends user, user is helped to mediate unhealthy emotion.
The embodiment of the method for the present invention is corresponded and is supplemented with the system embodiment of the present invention, and method of the invention is implemented
The technical detail of example can equally be well applied to the system embodiment of the present invention, repeats, repeats no more to reduce.
In addition, in the present invention, if be realized in the form of SFU software functional unit and as the sale of independent product or
In use, can be stored in a computer read/write memory medium.Based on this understanding, the technical solution sheet of the application
In matter in other words the part to the contribution of the prior art or the technical solution in whole or in part can be with software product
Form embodies, which is stored in a storage medium, including some instructions are used so that a meter
It calculates machine equipment (but personal computer, server either network equipment) or processor executes the application embodiment method
All or part of step.And storage above-mentioned is worth bag data server, cloud server, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), mobile communication equipment or CD,
Or USB flash disk etc. is various can be with the medium of store code.
In addition, various embodiments of the present invention can be combined randomly, as long as it is without prejudice to originally
The thought of invention equally should be considered as the content that the present invention is invented.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
God and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of method from social object to user that recommending, which is characterized in that including：
S100 obtains the expression information of user；
S200 generates recommendation according to the communication score information in the historical communication record storehouse of the expression information and the user
Social object；
The social object of the recommendation is presented to the user by S300, and establishes social in real time lead to the social object of the recommendation
Letter.
2. a kind of method from social object to user that recommending according to claim 1, which is characterized in that the historical communication
Record storehouse also includes the time of each historical communication；The step S200 is logical according to the history of the expression information and the user
Believe that the communication score information in record storehouse, the social object for generating recommendation specifically include：
S210 is when the expression information is happy or normal, according to the communication score information to the historical communication record storehouse
In communications records be ranked up, the current farthest communication object of the forward and historical communication time gap of sorting is as recommendation society
Hand over object.
3. a kind of method from social object to user that recommending according to claim 1, which is characterized in that the historical communication
Record storehouse also includes the initial mood of each historical communication；The step S200 is gone through according to the expression information and the user's
Communication score information in history communications records library, the social object for generating recommendation specifically include：
S220 when the expression information be it is sad or angry when, searched in the historical communication record storehouse initial mood with it is described
Expression information is identical and the highest communications records of scoring, using the corresponding communication object of the communications records as recommending social activity right
As.
4. according to a kind of method from social object to user that recommending of claim 1-3 any one of them, which is characterized in that in institute
Stating step S300 further includes later：
S400 carries out real-time state monitoring to the real-time social communication；
The communication that S500 calculates this real-time social communication according to the data monitored in real time is scored；
S600 records the communication scoring update of this real-time social communication record and the social communication to the historical communication
Library.
5. a kind of method from social object to user that recommending according to claim 4, which is characterized in that the real-time monitoring
Data include：Mood value, converse total duration, the mood of user's t moment when communicating mood value when starting, sign off
Voice duration in value, user's communication；The step S500 calculates this real-time social activity according to the data monitored in real time and leads to
The scoring of letter specifically includes：
S510 calculates the communication of this real-time social communication of the user by following equation according to the data monitored in real time
Scoring：
Wherein, S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is the time
Variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；M (t) is
The mood value of user described in t moment in communication process；P is the percentages shared by the voice of user in this communication；K1、
K2 is the proportionality coefficient constant of positive value.
6. a kind of system from social object to user that recommending, which is characterized in that including：
Expression acquisition module, the expression information for obtaining user；
Historical communication record storehouse, the historical communication for storing the user record, and the historical communication record is commented comprising communication
Divide information；
The historical communication of social analysis module, expression information and the user for being obtained according to the expression acquisition module is remembered
The communication score information in library is recorded, the social object of recommendation is generated；
Social recommendation module, the social object for the recommendation that the social analysis module generates to be presented to the user, and and
The social object of the recommendation establishes real-time social communication.
7. a kind of system from social object to user that recommending according to claim 6, which is characterized in that the historical communication
Record storehouse also records the time for having each historical communication；The social activity analysis module specifically includes：
Expression analysis submodule, the expression information for analyzing the user that the expression acquisition module obtains；
Sorting sub-module, when the expression information for the user described in the Expression analysis submodule block analysis is happy or normal,
The communications records in the historical communication record storehouse are ranked up according to institute's scoring information；
Object generates submodule, for the sorting sub-module to sort the current farthest use of forward and historical communication time gap
Family is as the social object of recommendation.
8. a kind of system from social object to user that recommending according to claim 7, which is characterized in that the historical communication
Record storehouse also records the initial mood for having each historical communication；It is described social activity analysis module further include：
Submodule is searched, when the expression information for the user described in the Expression analysis submodule block analysis is sad or angry,
Search that initial mood is identical as the expression information and the highest communications records of scoring in the historical communication record storehouse, it will
The corresponding communication object of the communications records is as recommending social object.
9. according to a kind of system from social object to user that recommending of claim 6-8 any one of them, which is characterized in that also wrap
It includes：
Communication monitoring module, for carrying out real-time state monitoring to the real-time social communication；
Grading module is communicated, the data for being monitored in real time according to the communication monitoring module calculate this real-time social communication
Communication scoring；
Module is updated storage, for updating the communication scoring of this real-time social communication record and the social communication to described
In historical communication record storehouse.
10. a kind of system from social object to user that recommending according to claim 9, which is characterized in that the communication prison
Surveying the data that module monitors in real time includes：Communication start when initial mood value, sign off when mood value, call it is total when
Voice duration in length, the mood value of user's t moment, user's communication；The communication grading module is in real time monitored according to described
The scoring that data calculate this real-time social communication specifically includes：
The data that the communication grading module monitors in real time according to the communication monitoring module, the use is calculated by following equation
The communication of this real-time social communication of family is scored：
Wherein：S is that the communication of this real-time social communication is scored；B is basic relation value；T is total air time；T is the time
Variable；The mood value of user when M2 is sign off；M1 is the initial mood value of user when communication starts；M (t) is
The mood value of user described in t moment in communication process；P is the percentages shared by the voice of user in this communication；K1、
K2 is the proportionality coefficient constant of positive value.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810295820.6A CN108494669A (en) | 2018-04-03 | 2018-04-03 | A kind of method and system from social object to user that recommending |
| PCT/CN2018/106781 WO2019192148A1 (en) | 2018-04-03 | 2018-09-20 | Method and system for recommending social object to user |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810295820.6A CN108494669A (en) | 2018-04-03 | 2018-04-03 | A kind of method and system from social object to user that recommending |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108494669A true CN108494669A (en) | 2018-09-04 |
Family
ID=63318123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810295820.6A Pending CN108494669A (en) | 2018-04-03 | 2018-04-03 | A kind of method and system from social object to user that recommending |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108494669A (en) |
| WO (1) | WO2019192148A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019192148A1 (en) * | 2018-04-03 | 2019-10-10 | 南京阿凡达机器人科技有限公司 | Method and system for recommending social object to user |
| CN112380451A (en) * | 2020-12-04 | 2021-02-19 | 江苏科技大学 | Favorite content recommendation method based on big data |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116796079B (en) * | 2023-06-30 | 2024-07-12 | 深圳市爱彼利科技有限公司 | Data processing method and device for social evaluation |
| CN117931958B (en) * | 2024-03-21 | 2024-06-18 | 山东山科世鑫科技有限公司 | Data warehouse management system and method based on artificial intelligence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905288A (en) * | 2012-12-26 | 2014-07-02 | 中国移动通信集团浙江有限公司 | Friend recommendation method and device |
| CN104917869A (en) * | 2014-03-10 | 2015-09-16 | 北京三星通信技术研究有限公司 | Mobile terminal, and contact information processing method and device for same |
| CN105635393A (en) * | 2014-10-30 | 2016-06-01 | 乐视致新电子科技（天津）有限公司 | Address book processing method and device |
| US9858316B1 (en) * | 2016-04-29 | 2018-01-02 | Rich Media Ventures, Llc | Mood-based settings |
| CN107734123A (en) * | 2016-08-12 | 2018-02-23 | 中兴通讯股份有限公司 | A kind of contact sequencing method and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104660770A (en) * | 2013-11-21 | 2015-05-27 | 中兴通讯股份有限公司 | Method and device for sequencing contact persons |
| CN105159979A (en) * | 2015-08-27 | 2015-12-16 | 广东小天才科技有限公司 | friend recommendation method and device |
| CN108494669A (en) * | 2018-04-03 | 2018-09-04 | 南京阿凡达机器人科技有限公司 | A kind of method and system from social object to user that recommending |
-
2018
- 2018-04-03 CN CN201810295820.6A patent/CN108494669A/en active Pending
- 2018-09-20 WO PCT/CN2018/106781 patent/WO2019192148A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905288A (en) * | 2012-12-26 | 2014-07-02 | 中国移动通信集团浙江有限公司 | Friend recommendation method and device |
| CN104917869A (en) * | 2014-03-10 | 2015-09-16 | 北京三星通信技术研究有限公司 | Mobile terminal, and contact information processing method and device for same |
| CN105635393A (en) * | 2014-10-30 | 2016-06-01 | 乐视致新电子科技（天津）有限公司 | Address book processing method and device |
| US9858316B1 (en) * | 2016-04-29 | 2018-01-02 | Rich Media Ventures, Llc | Mood-based settings |
| CN107734123A (en) * | 2016-08-12 | 2018-02-23 | 中兴通讯股份有限公司 | A kind of contact sequencing method and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019192148A1 (en) * | 2018-04-03 | 2019-10-10 | 南京阿凡达机器人科技有限公司 | Method and system for recommending social object to user |
| CN112380451A (en) * | 2020-12-04 | 2021-02-19 | 江苏科技大学 | Favorite content recommendation method based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019192148A1 (en) | 2019-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108494669A (en) | A kind of method and system from social object to user that recommending | |
| US20210407520A1 (en) | Detecting user identity in shared audio source contexts | |
| US20180314959A1 (en) | Cognitive music selection system and method | |
| CN109547814A (en) | Video recommendation method, device, server and storage medium | |
| CN114556354A (en) | Automatically determining and presenting personalized action items from an event | |
| US20180101776A1 (en) | Extracting An Emotional State From Device Data | |
| US9519863B2 (en) | Method and apparatus for a predictive tracking device | |
| JP5394118B2 (en) | Behavior analysis device and call center system | |
| CN107562789A (en) | Knowledge base problem update method, customer service robot and readable storage medium storing program for executing | |
| WO2019032996A1 (en) | Oral communication device and computing architecture for processing data and outputting user feedback, and related methods | |
| CN108804609A (en) | Song recommendation method and device | |
| CN107038217A (en) | Love and marriage object matching method and device | |
| CN112364234B (en) | Automatic grouping system for online discussion | |
| CN102654859A (en) | Method and system for recommending songs | |
| CN113748439A (en) | Prediction of successful quotient for motion pictures | |
| EP3200188A1 (en) | Computer implemented methods for assessing a disease through voice analysis and computer programs thereof | |
| US20190371344A1 (en) | Apparatus and method for predicting/recognizing occurrence of personal concerned context | |
| Yu et al. | Inferring user profile attributes from multidimensional mobile phone sensory data | |
| CN110210194A (en) | Electronic contract display methods, device, electronic equipment and storage medium | |
| CN109074397A (en) | Information processing system and information processing method | |
| US10888783B2 (en) | Dynamic modification of audio playback in games | |
| KR102149371B1 (en) | Method and system for matching workforce based on video | |
| CN115798513A (en) | Talent expression management method, system and computer readable storage medium | |
| CN112836138A (en) | User recommendation method and device | |
| WO2020230589A1 (en) | Information processing device, information processing method, and information processing program |
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 |
Application publication date: 20180904 |
|
| RJ01 | Rejection of invention patent application after publication |