CN109446378A - Information recommendation method, Sentiment orientation determine method and device and electronic equipment - Google Patents
Information recommendation method, Sentiment orientation determine method and device and electronic equipment Download PDFInfo
- Publication number
- CN109446378A CN109446378A CN201811325282.7A CN201811325282A CN109446378A CN 109446378 A CN109446378 A CN 109446378A CN 201811325282 A CN201811325282 A CN 201811325282A CN 109446378 A CN109446378 A CN 109446378A
- Authority
- CN
- China
- Prior art keywords
- sentiment orientation
- text
- target user
- interaction text
- emotion
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Abstract
The embodiment of the invention provides a kind of information recommendation method, Sentiment orientations to determine method and device and electronic equipment.This method comprises: obtaining the interaction text of target user；Using preset text trend analysis strategy, interaction text is analyzed, determines the Sentiment orientation of the interaction text；According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined；According to the Sentiment orientation of the target user, will information corresponding with the Sentiment orientation, and recommend target user.As it can be seen that, by analyzing the Sentiment orientation of target user, accurately determining the Sentiment orientation of target user using the embodiment of the present invention, carrying out information recommendation, to form a benign interaction, improve user feeling state, stimulate user activity.
Description
Technical field
The technical field of recommendation of personalized information the present invention relates to computer network based on user information, more particularly to
The determination method and device and electronic equipment of a kind of information recommendation method, device and user feeling tendency.
Background technique
With the explosive growth of information resources, the record for the information that many websites can be watched based on user, individual character
Change is that user recommends relevant information.
At present, it can be common that the personalized recommendation algorithm drawn a portrait based on content and user.For example, content-based recommendation is calculated
Method, the proposed algorithm based on correlation rule and Collaborative Filtering Recommendation Algorithm.
The algorithm of personalized recommendation based on content and user's portrait is mainly based upon to the existing use of information user
The record of data information recommends the relevant information content according to the feature association between data information for information user.Example
Such as, video user often sees children's video between winter and summer vacations, the algorithm mould based on the personalized recommendation that content and user are drawn a portrait
Type just sticks the label that there are children in family to the user, wherein user's portrait of the user is that family has children, likes seeing that children regard
Frequently.Then, the algorithm model for the personalized recommendation drawn a portrait based on content and user is recommended according to above-mentioned analysis for the user more
It is suitble to the information contents such as the cartoon of children's viewing.
Obviously, the algorithm for the personalized recommendation drawn a portrait based on content and user, is based only on the content of text of data information
Information be that user recommends associated with the content that it was watched information, there is no emotion of the consideration user in viewing information
Tendency is positive, neutral or passive Sentiment orientation.
Summary of the invention
The embodiment of the present invention be designed to provide a kind of information recommendation method and Sentiment orientation determine method and device and
Electronic equipment carries out information recommendation to realize the Sentiment orientation based on target user.
Specific technical solution is as follows:
In a first aspect, a kind of information recommendation method is provided, this method comprises:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
Sentiment orientation；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined；
According to the Sentiment orientation of the target user, information corresponding with the Sentiment orientation is obtained, and is recommended described
Target user.
Further, described to use preset text trend analysis strategy, the interaction text is analyzed, determines institute
State interaction text Sentiment orientation the step of, may include:
The interaction text is input to preset Sentiment orientation analysis model one by one；
Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.
It is further, described after the step of obtaining the interaction text of file destination, comprising:
Obtain the generation time of the interaction text；
The Sentiment orientation according to the interaction text, the step of determining the Sentiment orientation of the target user, comprising:
According to preset weight distribution strategy, the corresponding weight of the interaction text is determined, the interaction text is corresponding
The generation time that weight interacts text with this is positively correlated；
The interaction text pair is determined based on the Sentiment orientation of the interaction text according to preset emotion radix mapping table
The emotion radix answered；
According to the corresponding weight of the interaction text, the corresponding emotion radix of the interaction text is weighted, generates and uses
In the assessed value for the Sentiment orientation for determining the target user；
According to assessed value generated, the Sentiment orientation of the target user is determined.
Further, the interaction text of the positive emotion tendency in the preset emotion radix mapping table is corresponding actively
Emotion radix is positive number, and the corresponding Negative Affect radix of interaction text of Negative Affect tendency is negative, neutral Sentiment orientation
Interacting the corresponding neutral emotion radix of text is zero；And the positive emotion radix is added after being averaging with Negative Affect radix and counts
Obtained numerical value is equal with the neutral emotion radix；
It is described according to assessed value generated, the step of determining the Sentiment orientation of the target user, may include:
If assessed value generated is greater than the first preset threshold, the Sentiment orientation of the target user is positive；
If assessed value generated is passiveness less than the second preset threshold, the Sentiment orientation of the target user；
If assessed value generated is greater than or equal to second preset threshold, and assessed value generated is less than or equal to
First preset threshold, then the Sentiment orientation of the target user is neutrality；Wherein, first preset threshold is positive number,
Second preset threshold is negative.
Further, the Sentiment orientation according to the target user obtains information corresponding with the Sentiment orientation,
And the step of recommending the target user, may include:
According to the Sentiment orientation of the target user, when the Sentiment orientation of the target user is passive, obtain preparatory
It is identified as positive information, recommends target user；
When the Sentiment orientation of the target user is positive or neutral, obtain relevant to target user's watched information
Information recommends target user.
Further, it is described obtain target user interaction text the step of, may include:
Obtain all interaction texts of the target user in the preset duration before current time.
Second aspect, provides a kind of information recommending apparatus, and described device includes:
It interacts text and obtains module, for obtaining the interaction text of target user；
Text analysis model is interacted, for using preset text trend analysis strategy, the interaction text is divided
Analysis determines the Sentiment orientation of the interaction text；The Sentiment orientation includes: positive, neutral or passive；
Target user's Sentiment orientation determining module determines the target for the Sentiment orientation according to the interaction text
The Sentiment orientation of user；
Information recommendation module, for the Sentiment orientation according to the target user, acquisition it is corresponding with the Sentiment orientation
Information, and recommend the target user.
Further, the interaction text analysis model may include:
Text input submodule is interacted, analyzes mould for the interaction text to be input to preset Sentiment orientation one by one
Type；
Sentiment orientation receiving submodule is inclined for receiving the emotion of every interaction text of Sentiment orientation analysis model return
To.
Further, described device can also include:
Text generation time acquisition module is interacted, for obtaining the generation time of the interaction text；
Target user's Sentiment orientation determining module may include:
Interaction text weight determines submodule, for determining the interaction text pair according to preset weight distribution strategy
The weight answered, the generation time positive correlation for interacting the corresponding weight of text and interacting text with this；
Emotion radix determines submodule, is used for the feelings according to preset emotion radix mapping table, based on the interaction text
Sense tendency, determines the corresponding emotion radix of the interaction text；
Assessed value generates submodule, is used for according to the corresponding weight of the interaction text, to the corresponding feelings of interaction text
Sense radix is weighted, and generates the assessed value for determining the Sentiment orientation of the target user；
Sentiment orientation determines submodule, for determining the Sentiment orientation of the target user according to assessed value generated.
Further, the interaction text of the positive emotion tendency in the preset emotion radix mapping table is corresponding actively
Emotion radix is positive number, and the corresponding Negative Affect radix of interaction text of Negative Affect tendency is negative, neutral Sentiment orientation
Interacting the corresponding neutral emotion radix of text is zero；And the positive emotion radix is added after being averaging with Negative Affect radix and counts
Obtained numerical value is equal with the neutral emotion radix；
The Sentiment orientation determines submodule, may include:
Positive emotion is inclined to determination unit, if being greater than the first preset threshold, the target for assessed value generated
The Sentiment orientation of user is positive；
Negative Affect is inclined to determination unit, if for assessed value generated less than the second preset threshold, the target
The Sentiment orientation of user is passiveness；
Neutral Sentiment orientation determination unit, if being greater than or equal to second preset threshold for assessed value generated,
And assessed value generated is less than or equal to first preset threshold, then the Sentiment orientation of the target user is neutrality；Its
In, first preset threshold is positive number, and second preset threshold is negative.
Further, the information recommendation module may include:
Positive information recommendation submodule, for the Sentiment orientation according to the target user, when the feelings of the target user
When sense tendency is passive, acquisition is identified as positive information in advance, recommends target user；
Relevant information recommend submodule, for when the Sentiment orientation of the target user be it is positive or neutral when, obtain with
The relevant information of target user's watched information, recommends target user.
Further, the interaction text obtains module, may include:
It interacts text and obtains submodule, for obtaining all interactions of the target user in the preset duration before current time
Text.
The third aspect provides a kind of determination method of Sentiment orientation, which comprises
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
Sentiment orientation；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined.
Fourth aspect provides a kind of determining device of Sentiment orientation, which is characterized in that described device includes:
It interacts text and obtains module, for obtaining the interaction text of target user；
Text analysis model is interacted, for using preset text trend analysis strategy, the interaction text is divided
Analysis determines the Sentiment orientation of the interaction text；The Sentiment orientation includes: positive, neutral or passive；
Target user's Sentiment orientation determining module determines the target for the Sentiment orientation according to the interaction text
The Sentiment orientation of user.
The specific implementation for the device that each step and fourth aspect of method about third aspect offer provide can be joined
That sees above- mentioned information recommended method introduces process, and this will not be repeated here.
5th aspect, a kind of electronic equipment, which is characterized in that total including processor, communication interface, memory and communication
Line, wherein processor, communication interface, memory complete mutual communication by communication bus；
Memory, for storing computer program；
Processor when for executing the program stored on memory, realizes the information recommendation being inclined to based on user feeling
Method and step:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
Sentiment orientation；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined；
According to the Sentiment orientation of the target user, information corresponding with the Sentiment orientation is obtained, and is recommended described
Target user.
Or the determination method and step of Sentiment orientation may be implemented:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
Sentiment orientation；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined.
6th aspect, it is described computer-readable to deposit the embodiment of the invention also provides a kind of computer readable storage medium
Computer program is stored in storage media, the computer program realizes any of the above-described information recommendation method when being executed by processor
The step of determining method with Sentiment orientation.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method, apparatus and electronic equipment analyze the feelings of the every interaction text of target user by preset text trend analysis strategy
Sense tendency, determines the Sentiment orientation of target user；According to the Sentiment orientation of the target user, obtain and the Sentiment orientation
Corresponding information, and target user is recommended, the Sentiment orientation based on target user is realized, information recommendation is carried out.
Certainly, implement any of the products of the present invention or method it is not absolutely required at the same reach all the above excellent
Point.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of information recommendation method provided in an embodiment of the present invention；
Fig. 2 is a kind of structural schematic diagram of information recommending apparatus provided in an embodiment of the present invention；
Fig. 3 is a kind of flow diagram of the determination method of Sentiment orientation provided in an embodiment of the present invention；
Fig. 4 is a kind of structural schematic diagram of the determining device of Sentiment orientation provided in an embodiment of the present invention；
Fig. 5 is a kind of electronic equipment structural schematic diagram provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention provides a kind of information recommendation method, Sentiment orientations to determine method and device and electronic equipment,
The Sentiment orientation based on target user is realized, information recommendation is carried out.It is described in detail separately below.
Information recommendation method provided in an embodiment of the present invention is introduced first below.
Shown in Figure 1, the embodiment of the invention provides a kind of information recommendation methods, method includes the following steps:
S110: the interaction text of target user is obtained；
It should be noted that more and more users after use information, can provide platform in website or other information
On, oneself evaluation to article, data information or webpage is delivered by online rating system, the emotion for expressing oneself is known from experience.On
These information are stated, interaction text is referred to as.
Specifically, the interaction text of above-mentioned acquisition target user, which can be, obtains target user's presetting before current time
All interaction texts in duration.Under normal circumstances, when which can be for default before current time of target user
Comment information and/or chat message in length.Certainly, the text thumbed up by target user can also be used as the mutual of the target user
Dynamic text.
S120: using preset text trend analysis strategy, analyzes interaction text, determines the emotion of interaction text
Tendency；
It should be noted that all interaction texts of the target user obtained in S110, all embody target use indirectly
Certain Sentiment orientations at family；Meanwhile these Sentiment orientations of target user will also tend to embody the article of this kind of quality, data letter
The influence that breath or webpage generate target user.In general, the Sentiment orientation of target user may include: positive, neutral or passive.
Specifically, after obtaining the interaction text of above-mentioned target user, by specific text trend analysis strategy,
It can obtain the Sentiment orientation of above-mentioned interaction text.In the embodiment of the present invention, target user can be analyzed in the following way
Interaction text Sentiment orientation.
In a kind of optional embodiment, above-mentioned interaction text is input to preset Sentiment orientation one by one and analyzes mould
Type；Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.
Specifically, the interaction text of above-mentioned acquisition can be input to preset Sentiment orientation analysis model one by one, and connect
Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.Wherein, above-mentioned preset Sentiment orientation analyzes mould
Type can be Sentiment orientation analysis model of any open platform for Sentiment orientation analysis, for example, certain search website AI is open
The Sentiment orientation analysis model of platform.For example, being inputted when in the Sentiment orientation analysis model in certain search website AI open platform
Interacting text is " I am good happy today ", certain search website AI open platform calls participle tool first, to above-mentioned input text
Then word after participle is input in Sentiment orientation analysis model by this progress word segmentation processing, obtain what above-mentioned participle obtained
The corresponding Sentiment orientation of each word, the Sentiment orientation summation of all words then segmented to above-mentioned interaction text, obtains
Sentiment orientation result to the interaction text is positive；Similarly, the Sentiment orientation of above-mentioned interaction text is also possible to as neutrality or disappears
Pole.
In another optional embodiment, the text analyzing tool based on sentiment dictionary can be used, determines interaction
The Sentiment orientation of text.Wherein, different emotions word and its corresponding Sentiment orientation are had recorded in sentiment dictionary.
Specifically, being divided by the participle tool in text analyzing tool the interaction text of the target user of acquisition
Word processing obtains the word of every interaction text；Such as, the interaction text of user is " I am very unhappy today ", by segmenting work
Tool carries out word segmentation processing to the interaction text, available word: " I ", " today ", " very ", " no " and " happiness ", according to pre-
If sentiment dictionary, can determine " happiness " be above-mentioned interaction text emotion word, and " happiness " corresponding Sentiment orientation be accumulate
Pole, analysis personnel based on practical experience, determine the Sentiment orientation of above-mentioned interaction text.Such as, analysis personnel repairing according to negative word
Decorations can make the feeling polarities of emotion word change and when odd-times occurs in negative word, indicate the experiences such as negative meaning, really
The corresponding Sentiment orientation of above-mentioned interaction text " I am very unhappy today " is made as passiveness.
Technology known in those skilled in the art is belonged to for above-mentioned preset text trend analysis strategy, here no longer
It repeats.
S130: according to the Sentiment orientation of interaction text, the Sentiment orientation of target user is determined；
It should be noted that since the data information that target user watches in different time sections is different, the interaction text of formation
This content is different, then the Sentiment orientation of target user may be closely related with the time.It is used so obtaining target in S110 step
When each item interaction text of family, the generation time of every interaction text can also be obtained simultaneously, wherein the generation time specifically can be with
For the system timestamp for interacting the text generation time, in the implementation that the embodiment of the present invention has, which can be with
It is used as the weight of every interaction text., can be successive according to the time in other specific embodiments, to distribute weight, with
Current time is closer, and weight is higher.
Specifically, according to the Sentiment orientation of above-mentioned interaction text, determining that the emotion of target user is inclined in the embodiment of the present invention
To the step of, may include:
Firstly, determining the corresponding weight of interaction text according to preset weight distribution strategy；Wherein, interaction text is corresponding
Weight interacted with this text the generation time be positively correlated.
Then, which is determined based on the Sentiment orientation of interaction text according to preset emotion radix mapping table
The emotion radix answered.
When implementing, the Sentiment orientation result of above-mentioned every interaction text can correspond to an emotion radix.Specifically,
In preset emotion radix mapping table, the corresponding positive emotion radix of interaction text of positive emotion tendency is positive number, passive
The corresponding Negative Affect radix of the interaction text of Sentiment orientation is negative, the corresponding middle disposition of the interaction text of neutral Sentiment orientation
Feeling radix is zero；And the positive emotion radix be added with Negative Affect radix after being averaging the numerical value that is calculated and it is described in
Disposition sense radix is equal.
For example, corresponding emotion radix can be with when the Sentiment orientation of the interaction text of above-mentioned target user is positive
It is 1；When the Sentiment orientation of the interaction text of target user is passive, corresponding emotion radix can be -1；When target is used
When the Sentiment orientation of the interaction text at family is neutral, corresponding emotion radix can be 0.Wherein, in embodiments of the present invention,
Above-mentioned (1,0, -1) is properly termed as to the emotion radix of interaction text.
It should be noted that the corresponding emotion radix of above-mentioned interaction text was not limited only to provide in the embodiment of the present invention
Positive number, zero and negative can also be any numerical value that can be used for identifying interaction text emotion tendency, here without limitation.
Later, according to the corresponding weight of interaction text, the corresponding emotion radix of the interaction text is weighted, generates and uses
In the assessed value for the Sentiment orientation for determining the target user.
Specifically, the corresponding weight of emotion cardinal sum of above-mentioned every interaction text can be used, it can by following formula
The assessed value of Sentiment orientation of the target user in preset duration is calculated.
Wherein, i is i-th bar interaction text of the target user in preset duration, and value can be 1 ..., and n, n are to obtain
The total number of interaction text of the target user taken in the preset duration, WiI-th corresponding interaction time of interaction text,
Its unit is second, BiFor emotion radix corresponding to i-th interaction text, C is the emotion for stating target user in preset duration
The assessed value of tendency.
Obviously, the assessed value of Sentiment orientation of the above-mentioned target user in preset duration be to the target user when default
The summation of the multiplied result of the emotion radix weight corresponding with the interaction text of every interaction text in length.
It, then can be in addition, after obtaining the assessed value of Sentiment orientation of the target user in preset duration in this step
Judge the Sentiment orientation of the user for positive, neutral or passiveness.
Specifically, if the assessed value of Sentiment orientation of the above-mentioned target user in preset duration is greater than the first default threshold
Value, then Sentiment orientation of the target user in preset duration is positive；If emotion of the above-mentioned target user in preset duration is inclined
To assessed value less than the second preset threshold, then Sentiment orientation of the target user in preset duration is passiveness；If above-mentioned target
The assessed value of Sentiment orientation of the user in preset duration is greater than or equal to the second preset threshold, and assessed value generated is less than
Or being equal to the first preset threshold, then Sentiment orientation of the target user in preset duration is neutrality.Wherein, the first preset threshold is
Positive number, the second preset threshold are negative.
It is clearer in order to describe, two examples are set forth below, explain in the preset duration by above-mentioned target user
The corresponding weight of every interaction text every interaction text of corresponding emotion cardinal sum determines point of the Sentiment orientation of target user
Analysis process.
Table 1 is target user Zhang San, in the period: in 0 minute and 0 second 0 point of on May 2nd, 2018 to 0 minute and 0 second 0 point of May 9
Interact text statistical form.
Table 1
Table 2 is target user Li Si, in the period: in 0 minute and 0 second 0 point of on May 2nd, 2018 to 0 minute and 0 second 0 point of May 9
Interact text statistical form.
Table 2
It is found that Zhang San is greater than in the assessed value of the Sentiment orientation of above-mentioned period for 1525413720 as shown in 1 table 2 of table
Zero positive integer then illustrates that user Zhang San in the Sentiment orientation of the period is positive；As shown in Table 2 it is found that Li Si is above-mentioned
The assessed value of the Sentiment orientation of period be -406012, then illustrate user Li Si the period Sentiment orientation for passiveness.
S140: according to the Sentiment orientation of target user, obtaining information corresponding with Sentiment orientation, and recommends target use
Family.
It should be noted that information corresponding with Sentiment orientation is by information labeled in advance, the labeled information
It can be handmarking, be also possible to through ad hoc approach, machine or mark mode known in those skilled in the art, this
Inventive embodiments are defined not to this.
In order to form a benign interaction atmosphere, improves the affective state of user, user activity is stimulated, in the reality having
It applies in example, when the Sentiment orientation of target user is passive, will can be identified as in advance positive information recommendation and be used to target
Family.
In addition, will can be identified as in advance positive or neutral when the Sentiment orientation of target user is positive or neutral
Information recommendation to target user, can also will information relevant to target user's watched information, recommend target user.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method, apparatus and electronic equipment analyze the feelings of the every interaction text of target user by preset text trend analysis strategy
Sense tendency, determines the Sentiment orientation of target user；It is will acquire with above-mentioned Sentiment orientation according to the Sentiment orientation of target user
Corresponding information recommendation realizes the Sentiment orientation for accurately determining target user, to carry out information recommendation to target user.
In addition, when the Sentiment orientation of target user is passive, will can be identified as to accumulate in advance in the embodiment having
The information recommendation of pole is capable of forming a benign interaction to target user, improves user feeling state, and stimulation user is active
Degree.
Corresponding to above method embodiment, the embodiment of the invention also provides a kind of information recommending apparatus, institute referring to fig. 2
Show, which includes:
It interacts text and obtains module 210, for obtaining the interaction text of target user；
Text analysis model 220 is interacted, for using preset text trend analysis strategy, interaction text is divided
Analysis determines the Sentiment orientation of interaction text；Sentiment orientation includes: positive, neutral or passive；
Target user's Sentiment orientation determining module 230 determines target user's for the Sentiment orientation according to interaction text
Sentiment orientation；
Information recommendation module 240 obtains information corresponding with Sentiment orientation for the Sentiment orientation according to target user,
And recommend target user.
In embodiments of the present invention, above-mentioned interaction text analysis model may include:
Text input submodule is interacted, is input to preset Sentiment orientation analysis model one by one for text will to be interacted；
Sentiment orientation receiving submodule is inclined for receiving the emotion of every interaction text of Sentiment orientation analysis model return
To.
In embodiments of the present invention, above-mentioned apparatus can also include:
Text generation time acquisition module is interacted, for obtaining the generation time of interaction text；
Target user's Sentiment orientation determining module may include:
Interaction text weight determines submodule, for determining that interaction text is corresponding according to preset weight distribution strategy
Weight, the generation time that the corresponding weight of interaction text interacts text with this are positively correlated；
Emotion radix determines submodule, for according to preset emotion radix mapping table, the emotion based on interaction text to be inclined
To determining the corresponding emotion radix of the interaction text；
Assessed value generates submodule, for according to the corresponding weight of interaction text, emotion base corresponding to the interaction text
Number is weighted, and generates the assessed value for determining the Sentiment orientation of target user；
Sentiment orientation determines submodule, for determining the Sentiment orientation of target user according to assessed value generated.
In embodiments of the present invention, the interaction text of the positive emotion tendency in preset emotion radix mapping table is corresponding
Positive emotion radix is positive number, and the corresponding Negative Affect radix of interaction text of Negative Affect tendency is negative, and neutral emotion is inclined
To the corresponding neutral emotion radix of interaction text be zero；And positive emotion radix is added after being averaging with Negative Affect radix and counts
Obtained numerical value is equal with neutral emotion radix；
Above-mentioned Sentiment orientation determines submodule, may include:
Positive emotion is inclined to determination unit, if being greater than the first preset threshold, target user for assessed value generated
Sentiment orientation be positive；
Negative Affect is inclined to determination unit, if for assessed value generated less than the second preset threshold, target user
Sentiment orientation be passiveness；
Neutral Sentiment orientation determination unit, if being greater than or equal to second preset threshold for assessed value generated,
And assessed value generated is less than or equal to first preset threshold, then the Sentiment orientation of target user is neutrality；Wherein, institute
Stating the first preset threshold is positive number, and second preset threshold is negative.
In embodiments of the present invention, above- mentioned information recommending module may include:
Positive information recommendation submodule, for the Sentiment orientation according to target user, when the Sentiment orientation of target user is
When passive, acquisition is identified as positive information in advance, recommends target user；
Relevant information recommend submodule, for when the Sentiment orientation of target user be it is positive or neutral when, obtain and target
The relevant information of user's watched information, recommends target user.
In embodiments of the present invention, above-mentioned interaction text obtains module, may include:
It interacts text and obtains submodule, for obtaining all interactions of the target user in the preset duration before current time
Text.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method and device and electronic equipment analyze the every interaction text of target user by preset text trend analysis method
Sentiment orientation determines the Sentiment orientation of target user；According to the Sentiment orientation of target user, what be will acquire inclines with above-mentioned emotion
To corresponding information recommendation to target user, the Sentiment orientation for accurately determining target user is realized, is pushed away to carry out information
It recommends.
In addition, when the Sentiment orientation of target user is passive, will can be identified as to accumulate in advance in the embodiment having
The information recommendation of pole is capable of forming a benign interaction to target user, improves user feeling state, and stimulation user is active
Degree.
Shown in Figure 3, the embodiment of the invention also provides a kind of determination methods of Sentiment orientation, this method comprises:
S310: the interaction text of target user is obtained；
S320: using preset text trend analysis strategy, analyzes interaction text, determines the emotion of interaction text
Tendency；
S330: according to the Sentiment orientation of interaction text, the Sentiment orientation of target user is determined.
In embodiments of the present invention, above-mentioned to use preset text trend analysis strategy, interaction text is analyzed, really
Surely interact text Sentiment orientation the step of, may include:
Interaction text is input to preset Sentiment orientation analysis model one by one；
Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.
In embodiments of the present invention, above-mentioned after the step of obtaining the interaction text of file destination, may include:
Obtain the generation time of interaction text；
According to the Sentiment orientation of interaction text, the step of determining the Sentiment orientation of target user, may include:
According to preset weight distribution strategy, determines the corresponding weight of interaction text, interact the corresponding weight of text and be somebody's turn to do
The generation time for interacting text is positively correlated；
Determine that the interaction text is corresponding based on the Sentiment orientation of interaction text according to preset emotion radix mapping table
Emotion radix；
According to the corresponding weight of interaction text, the corresponding emotion radix of the interaction text is weighted, is generated for true
Set the goal user Sentiment orientation assessed value；
According to assessed value generated, the Sentiment orientation of target user is determined.
In embodiments of the present invention, the interaction text pair of the positive emotion tendency in above-mentioned preset emotion radix mapping table
The positive emotion radix answered is positive number, and the corresponding Negative Affect radix of interaction text of Negative Affect tendency is negative, middle disposition
The corresponding neutral emotion radix of interaction text for feeling tendency is zero；And positive emotion radix is added averaging with Negative Affect radix
The numerical value being calculated afterwards is equal with neutral emotion radix；
According to assessed value generated, the step of determining the Sentiment orientation of target user, may include:
If assessed value generated is greater than the first preset threshold, the Sentiment orientation of target user is positive；
If assessed value generated is passiveness less than the second preset threshold, the Sentiment orientation of target user；
If assessed value generated is greater than or equal to second preset threshold, and assessed value generated is less than or equal to
First preset threshold, then the Sentiment orientation of target user is neutrality；Wherein, first preset threshold is positive number, described
Second preset threshold is negative.
In embodiments of the present invention, the step of interaction text of above-mentioned acquisition target user may include:
Obtain all interaction texts of the target user in the preset duration before current time.
Specific implementation and relevant explanation content about each step of this method may refer to above-mentioned method shown in FIG. 1
Embodiment, this will not be repeated here.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method and device and electronic equipment analyze the every interaction text of target user by preset text trend analysis method
Sentiment orientation determines the Sentiment orientation of target user, realizes the Sentiment orientation for accurately determining target user.
Shown in Figure 4, the embodiment of the invention also provides a kind of determining device of Sentiment orientation, which includes:
It interacts text and obtains module 410, for obtaining the interaction text of target user；
Text analysis model 420 is interacted, for using preset text trend analysis strategy, interaction text is divided
Analysis determines the Sentiment orientation of interaction text；
Target user's Sentiment orientation determining module 430 determines that the target is used for the Sentiment orientation according to interaction text
The Sentiment orientation at family.
In embodiments of the present invention, above-mentioned interaction text analysis model may include:
Text input submodule is interacted, is input to preset Sentiment orientation analysis model one by one for text will to be interacted；
Sentiment orientation receiving submodule is inclined for receiving the emotion of every interaction text of Sentiment orientation analysis model return
To.
In embodiments of the present invention, above-mentioned apparatus further include:
Text generation time acquisition module is interacted, for obtaining the generation time of interaction text；
Target user's Sentiment orientation determining module may include:
Interaction text weight determines submodule, for determining that interaction text is corresponding according to preset weight distribution strategy
Weight, the generation time that the corresponding weight of interaction text interacts text with this are positively correlated；
Emotion radix determines submodule, for according to preset emotion radix mapping table, the emotion based on interaction text to be inclined
To determining the corresponding emotion radix of the interaction text；
Assessed value generates submodule, for according to the corresponding weight of interaction text, emotion base corresponding to the interaction text
Number is weighted, and generates the assessed value for determining the Sentiment orientation of target user；
Sentiment orientation determines submodule, for determining the Sentiment orientation of target user according to assessed value generated.
In embodiments of the present invention, the interaction text pair of the positive emotion tendency in above-mentioned preset emotion radix mapping table
The positive emotion radix answered is positive number, and the corresponding Negative Affect radix of interaction text of Negative Affect tendency is negative, middle disposition
The corresponding neutral emotion radix of interaction text for feeling tendency is zero；And positive emotion radix is added averaging with Negative Affect radix
The numerical value being calculated afterwards is equal with the neutral emotion radix；
Sentiment orientation determines submodule, may include:
Positive emotion is inclined to determination unit, if being greater than the first preset threshold, target user for assessed value generated
Sentiment orientation be positive；
Negative Affect is inclined to determination unit, if for assessed value generated less than the second preset threshold, target user
Sentiment orientation be passiveness；
Neutral Sentiment orientation determination unit, if being greater than or equal to the second preset threshold, and institute for assessed value generated
The assessed value of generation is less than or equal to first preset threshold, then the Sentiment orientation of target user is neutrality；Wherein, first is pre-
If threshold value is positive number, the second preset threshold is negative.
In embodiments of the present invention, above-mentioned interaction text obtains module, may include:
It interacts text and obtains submodule, for obtaining all interactions of the target user in the preset duration before current time
Text.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method and device and electronic equipment analyze the every interaction text of target user by preset text trend analysis method
Sentiment orientation determines the Sentiment orientation of target user, realizes the Sentiment orientation for accurately determining target user.
Shown in Figure 5, the embodiment of the invention provides a kind of electronic equipment, including processor, communication interface, memory
And communication bus, wherein processor, communication interface, memory complete mutual communication by communication bus；
Memory, for storing computer program；
Processor when for executing the program stored on memory, realizes information recommendation method step:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, interaction text is analyzed, determines the Sentiment orientation of interaction text；
Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of interaction text, the Sentiment orientation of target user is determined；
According to the Sentiment orientation of target user, information corresponding with Sentiment orientation is obtained, and recommends target user.
Or the step of realizing the determination method of following Sentiment orientation:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, interaction text is analyzed, determines the Sentiment orientation of interaction text；
Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of interaction text, the Sentiment orientation of target user is determined.
Specific implementation and relevant explanation content about each step of this method may refer to shown in above-mentioned Fig. 1 and Fig. 3
Embodiment of the method, this will not be repeated here.
As seen from the above technical solutions, a kind of information recommendation method provided in an embodiment of the present invention, Sentiment orientation determine
Method and device and electronic equipment analyze the every interaction text of target user by preset text trend analysis method
Sentiment orientation determines the Sentiment orientation of target user；According to the Sentiment orientation of target user, what be will acquire inclines with above-mentioned emotion
To corresponding information recommendation to target user, the Sentiment orientation for accurately determining target user is realized, is pushed away to carry out information
It recommends.
In addition, when the Sentiment orientation of target user is passive, will can be identified as to accumulate in advance in the embodiment having
The information recommendation of pole is capable of forming a benign interaction to target user, improves user feeling state, and stimulation user is active
Degree.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.；It can also be digital signal processor (Digital Signal
Processing, DSP), it is specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.In another embodiment provided by the invention, a kind of calculating is additionally provided
Machine readable storage medium storing program for executing is stored with computer program in the computer readable storage medium, and the computer program is by processor
The step of any of the above-described information recommendation method and Sentiment orientation determine method is realized when execution.
In another embodiment provided by the invention, a kind of computer program product comprising instruction is additionally provided, when it
When running on computers, so that computer executes any information recommended method and Sentiment orientation determination side in above-described embodiment
Method.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program
Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or
It partly generates according to process or function described in the embodiment of the present invention.The computer can be general purpose computer, dedicated meter
Calculation machine, computer network or other programmable devices.The computer instruction can store in computer readable storage medium
In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer
Instruction can pass through wired (such as coaxial cable, optical fiber, number from a web-site, computer, server or data center
User's line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or
Data center is transmitted.The computer readable storage medium can be any usable medium that computer can access or
It is comprising data storage devices such as one or more usable mediums integrated server, data centers.The usable medium can be with
It is magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (23)
1. a kind of information recommendation method, which is characterized in that the described method includes:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the emotion of the interaction text
Tendency；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined；
According to the Sentiment orientation of the target user, information corresponding with the Sentiment orientation is obtained, and recommends the target
User.
2. information recommendation method according to claim 1, which is characterized in that
It is described to use preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
The step of Sentiment orientation, comprising:
The interaction text is input to preset Sentiment orientation analysis model one by one；
Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.
3. information recommendation method according to claim 1, which is characterized in that described in the interaction text for obtaining file destination
The step of after, comprising:
Obtain the generation time of the interaction text；
The Sentiment orientation according to the interaction text, the step of determining the Sentiment orientation of the target user, comprising:
According to preset weight distribution strategy, the corresponding weight of the interaction text is determined, the corresponding weight of the interaction text
The generation time that text is interacted with this is positively correlated；
Determine that the interaction text is corresponding based on the Sentiment orientation of the interaction text according to preset emotion radix mapping table
Emotion radix；
According to the corresponding weight of the interaction text, the corresponding emotion radix of the interaction text is weighted, is generated for true
The assessed value of the Sentiment orientation of the fixed target user；
According to assessed value generated, the Sentiment orientation of the target user is determined.
4. according to information recommendation method as claimed in claim 3, which is characterized in that
The corresponding positive emotion radix of interaction text of positive emotion tendency in the preset emotion radix mapping table is positive
Number, the corresponding Negative Affect radix of interaction text of Negative Affect tendency are negative, and the interaction text of neutral Sentiment orientation is corresponding
Neutral emotion radix be zero；And the positive emotion radix is added the numerical value being calculated after being averaging with Negative Affect radix
It is equal with the neutral emotion radix；
It is described according to assessed value generated, the step of determining the Sentiment orientation of the target user, comprising:
If assessed value generated is greater than the first preset threshold, the Sentiment orientation of the target user is positive；
If assessed value generated is passiveness less than the second preset threshold, the Sentiment orientation of the target user；
If assessed value generated is greater than or equal to second preset threshold, and assessed value generated is less than or equal to described
First preset threshold, then the Sentiment orientation of the target user is neutrality；Wherein, first preset threshold is positive number, described
Second preset threshold is negative.
5. information recommendation method according to claim 1, which is characterized in that the emotion according to the target user is inclined
To, acquisition information corresponding with the Sentiment orientation, and the step of recommending the target user, comprising:
According to the Sentiment orientation of the target user, when the Sentiment orientation of the target user is passive, acquisition is marked in advance
Knowing is positive information, recommends target user；
When the Sentiment orientation of the target user is positive or neutral, letter relevant to target user's watched information is obtained
Breath, recommends target user.
6. -5 any information recommendation method according to claim 1, which is characterized in that
The step of interaction text for obtaining target user, comprising:
Obtain all interaction texts of the target user in the preset duration before current time.
7. a kind of information recommending apparatus, which is characterized in that described device includes:
It interacts text and obtains module, for obtaining the interaction text of target user；
Text analysis model is interacted, for using preset text trend analysis strategy, the interaction text is analyzed, really
The Sentiment orientation of the fixed interaction text；The Sentiment orientation includes: positive, neutral or passive；
Target user's Sentiment orientation determining module determines the target user for the Sentiment orientation according to the interaction text
Sentiment orientation；
Information recommendation module obtains information corresponding with the Sentiment orientation for the Sentiment orientation according to the target user,
And recommend the target user.
8. device according to claim 7, which is characterized in that the interaction text analysis model, comprising:
Text input submodule is interacted, for the interaction text to be input to preset Sentiment orientation analysis model one by one；
Sentiment orientation receiving submodule, the Sentiment orientation of the every interaction text for receiving the return of Sentiment orientation analysis model.
9. device according to claim 7, which is characterized in that described device further include:
Text generation time acquisition module is interacted, for obtaining the generation time of the interaction text；
Target user's Sentiment orientation determining module, comprising:
Interaction text weight determines submodule, for determining that the interaction text is corresponding according to preset weight distribution strategy
Weight, the generation time positive correlation for interacting the corresponding weight of text and interacting text with this；
Emotion radix determines submodule, for according to preset emotion radix mapping table, the emotion based on the interaction text to be inclined
To determining the corresponding emotion radix of the interaction text；
Assessed value generates submodule, for according to the corresponding weight of the interaction text, emotion base corresponding to the interaction text
Number is weighted, and generates the assessed value for determining the Sentiment orientation of the target user；
Sentiment orientation determines submodule, for determining the Sentiment orientation of the target user according to assessed value generated.
10. device according to claim 9, which is characterized in that
The corresponding positive emotion radix of interaction text of positive emotion tendency in the preset emotion radix mapping table is positive
Number, the corresponding Negative Affect radix of interaction text of Negative Affect tendency are negative, and the interaction text of neutral Sentiment orientation is corresponding
Neutral emotion radix be zero；And the positive emotion radix is added the numerical value being calculated after being averaging with Negative Affect radix
It is equal with the neutral emotion radix；
The Sentiment orientation determines submodule, comprising:
Positive emotion is inclined to determination unit, if being greater than the first preset threshold, the target user for assessed value generated
Sentiment orientation be positive；
Negative Affect is inclined to determination unit, if for assessed value generated less than the second preset threshold, the target user
Sentiment orientation be passiveness；
Neutral Sentiment orientation determination unit, if being greater than or equal to second preset threshold, and institute for assessed value generated
The assessed value of generation is less than or equal to first preset threshold, then the Sentiment orientation of the target user is neutrality；Wherein, institute
Stating the first preset threshold is positive number, and second preset threshold is negative.
11. device according to claim 7, which is characterized in that the information recommendation module, comprising:
Positive information recommendation submodule, for the Sentiment orientation according to the target user, when the emotion of the target user is inclined
To be passive when, acquisition is identified as positive information in advance, recommends target user；
Relevant information recommend submodule, for when the Sentiment orientation of the target user be it is positive or neutral when, obtain and target
The relevant information of user's watched information, recommends target user.
12. according to any device of claim 7-11, which is characterized in that
The interaction text obtains module, comprising:
It interacts text and obtains submodule, for obtaining all interactions text of the target user in the preset duration before current time
This.
13. a kind of determination method of Sentiment orientation, which is characterized in that the described method includes:
Obtain the interaction text of target user；
Using preset text trend analysis strategy, the interaction text is analyzed, determines the emotion of the interaction text
Tendency；The Sentiment orientation includes: positive, neutral or passive；
According to the Sentiment orientation of the interaction text, the Sentiment orientation of the target user is determined.
14. the determination method of Sentiment orientation according to claim 13, which is characterized in that
It is described to use preset text trend analysis strategy, the interaction text is analyzed, determines the interaction text
The step of Sentiment orientation, comprising:
The interaction text is input to preset Sentiment orientation analysis model one by one；
Receive the Sentiment orientation for the every interaction text that Sentiment orientation analysis model returns.
15. the determination method of user feeling tendency according to claim 13, which is characterized in that
It is described obtain file destination interaction text the step of after, comprising:
Obtain the generation time of the interaction text；
The Sentiment orientation according to the interaction text, the step of determining the Sentiment orientation of the target user, comprising:
According to preset weight distribution strategy, the corresponding weight of the interaction text is determined, the corresponding weight of the interaction text
The generation time that text is interacted with this is positively correlated；
Determine that the interaction text is corresponding based on the Sentiment orientation of the interaction text according to preset emotion radix mapping table
Emotion radix；
According to the corresponding weight of the interaction text, the corresponding emotion radix of the interaction text is weighted, is generated for true
The assessed value of the Sentiment orientation of the fixed target user；
According to assessed value generated, the Sentiment orientation of the target user is determined.
16. the determination method of Sentiment orientation according to claim 15, which is characterized in that
The corresponding positive emotion radix of interaction text of positive emotion tendency in the preset emotion radix mapping table is positive
Number, the corresponding Negative Affect radix of interaction text of Negative Affect tendency are negative, and the interaction text of neutral Sentiment orientation is corresponding
Neutral emotion radix be zero；And the positive emotion radix is added the numerical value being calculated after being averaging with Negative Affect radix
It is equal with the neutral emotion radix；
It is described according to assessed value generated, the step of determining the Sentiment orientation of the target user, comprising:
If assessed value generated is greater than the first preset threshold, the Sentiment orientation of the target user is positive；
If assessed value generated is passiveness less than the second preset threshold, the Sentiment orientation of the target user；
If assessed value generated is greater than or equal to second preset threshold, and assessed value generated is less than or equal to described
First preset threshold, then the Sentiment orientation of the target user is neutrality；Wherein, first preset threshold is positive number, described
Second preset threshold is negative.
17. the determination method of any Sentiment orientation of 3-16 according to claim 1, which is characterized in that
The step of interaction text for obtaining target user, comprising:
Obtain all interaction texts of the target user in the preset duration before current time.
18. a kind of determining device of Sentiment orientation, which is characterized in that described device includes:
It interacts text and obtains module, for obtaining the interaction text of target user；
Text analysis model is interacted, for using preset text trend analysis strategy, the interaction text is analyzed, really
The Sentiment orientation of the fixed interaction text；The Sentiment orientation includes: positive, neutral or passive；
Target user's Sentiment orientation determining module determines the target user for the Sentiment orientation according to the interaction text
Sentiment orientation.
19. device according to claim 18, which is characterized in that the interaction text analysis model, comprising:
Text input submodule is interacted, for the interaction text to be input to preset Sentiment orientation analysis model one by one；
Sentiment orientation receiving submodule, the Sentiment orientation of the every interaction text for receiving the return of Sentiment orientation analysis model.
20. device according to claim 18, which is characterized in that described device further include:
Text generation time acquisition module is interacted, for obtaining the generation time of the interaction text；
Target user's Sentiment orientation determining module, comprising:
Interaction text weight determines submodule, for determining that the interaction text is corresponding according to preset weight distribution strategy
Weight, the generation time positive correlation for interacting the corresponding weight of text and interacting text with this；
Emotion radix determines submodule, for according to preset emotion radix mapping table, the emotion based on the interaction text to be inclined
To determining the corresponding emotion radix of the interaction text；
Assessed value generates submodule, for according to the corresponding weight of the interaction text, emotion base corresponding to the interaction text
Number is weighted, and generates the assessed value for determining the Sentiment orientation of the target user；
Sentiment orientation determines submodule, for determining the Sentiment orientation of the target user according to assessed value generated.
21. according to the method for claim 20, which is characterized in that the positive feelings in the preset emotion radix mapping table
The corresponding positive emotion radix of interaction text of sense tendency is positive number, the corresponding Negative Affect of interaction text of Negative Affect tendency
Radix is negative, and the corresponding neutral emotion radix of the interaction text of neutral Sentiment orientation is zero；And the positive emotion radix with
It is equal with the neutral emotion radix that Negative Affect radix is added the numerical value being calculated after averaging；
The Sentiment orientation determines submodule, comprising:
Positive emotion is inclined to determination unit, if being greater than the first preset threshold, the target user for assessed value generated
Sentiment orientation be positive；
Negative Affect is inclined to determination unit, if for assessed value generated less than the second preset threshold, the target user
Sentiment orientation be passiveness；
Neutral Sentiment orientation determination unit, if being greater than or equal to second preset threshold, and institute for assessed value generated
The assessed value of generation is less than or equal to first preset threshold, then the Sentiment orientation of the target user is neutrality；Wherein, institute
Stating the first preset threshold is positive number, and second preset threshold is negative.
22. any device of 8-21 according to claim 1, which is characterized in that
The interaction text obtains module, comprising:
It interacts text and obtains submodule, for obtaining all interactions text of the target user in the preset duration before current time
This.
23. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing
Device, communication interface, memory complete mutual communication by communication bus；
Memory, for storing computer program；
Processor when for executing the program stored on memory, realizes that claim 1-6 or claim 13-17 is any
The method and step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325282.7A CN109446378A (en) | 2018-11-08 | 2018-11-08 | Information recommendation method, Sentiment orientation determine method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325282.7A CN109446378A (en) | 2018-11-08 | 2018-11-08 | Information recommendation method, Sentiment orientation determine method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109446378A true CN109446378A (en) | 2019-03-08 |
Family
ID=65551226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811325282.7A Pending CN109446378A (en) | 2018-11-08 | 2018-11-08 | Information recommendation method, Sentiment orientation determine method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109446378A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112218114A (en) * | 2019-07-10 | 2021-01-12 | 中国电信股份有限公司 | Video cache control method, device and computer readable storage medium |
| CN112906399A (en) * | 2021-02-20 | 2021-06-04 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for determining emotional state |
| CN115113781A (en) * | 2022-06-28 | 2022-09-27 | 广州博冠信息科技有限公司 | Interactive icon display method, device, medium and electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103699626A (en) * | 2013-12-20 | 2014-04-02 | 华南理工大学 | Method and system for analysing individual emotion tendency of microblog user |
| CN105095183A (en) * | 2014-05-22 | 2015-11-25 | 株式会社日立制作所 | Text emotional tendency determination method and system |
| CN105426381A (en) * | 2015-08-27 | 2016-03-23 | 浙江大学 | Music recommendation method based on emotional context of microblog |
| CN106649603A (en) * | 2016-11-25 | 2017-05-10 | 北京资采信息技术有限公司 | Webpage text data sentiment classification designated information push method |
| CN107103093A (en) * | 2017-05-16 | 2017-08-29 | 武汉大学 | A kind of short text based on user behavior and sentiment analysis recommends method and device |
| CN107992550A (en) * | 2017-11-28 | 2018-05-04 | 沈阳建筑大学 | A kind of network comment analysis method and system |
| US20180316635A1 (en) * | 2017-04-26 | 2018-11-01 | International Business Machines Corporation | Recommendation generator responsive to messaging content |
| CN108763210A (en) * | 2018-05-22 | 2018-11-06 | 华中科技大学 | A kind of sentiment analysis and forecasting system based on automated data collection |
-
2018
- 2018-11-08 CN CN201811325282.7A patent/CN109446378A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103699626A (en) * | 2013-12-20 | 2014-04-02 | 华南理工大学 | Method and system for analysing individual emotion tendency of microblog user |
| CN105095183A (en) * | 2014-05-22 | 2015-11-25 | 株式会社日立制作所 | Text emotional tendency determination method and system |
| CN105426381A (en) * | 2015-08-27 | 2016-03-23 | 浙江大学 | Music recommendation method based on emotional context of microblog |
| CN106649603A (en) * | 2016-11-25 | 2017-05-10 | 北京资采信息技术有限公司 | Webpage text data sentiment classification designated information push method |
| US20180316635A1 (en) * | 2017-04-26 | 2018-11-01 | International Business Machines Corporation | Recommendation generator responsive to messaging content |
| CN107103093A (en) * | 2017-05-16 | 2017-08-29 | 武汉大学 | A kind of short text based on user behavior and sentiment analysis recommends method and device |
| CN107992550A (en) * | 2017-11-28 | 2018-05-04 | 沈阳建筑大学 | A kind of network comment analysis method and system |
| CN108763210A (en) * | 2018-05-22 | 2018-11-06 | 华中科技大学 | A kind of sentiment analysis and forecasting system based on automated data collection |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112218114A (en) * | 2019-07-10 | 2021-01-12 | 中国电信股份有限公司 | Video cache control method, device and computer readable storage medium |
| CN112218114B (en) * | 2019-07-10 | 2023-01-03 | 中国电信股份有限公司 | Video cache control method, device and computer readable storage medium |
| CN112906399A (en) * | 2021-02-20 | 2021-06-04 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for determining emotional state |
| CN112906399B (en) * | 2021-02-20 | 2023-11-10 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for determining emotional state |
| CN115113781A (en) * | 2022-06-28 | 2022-09-27 | 广州博冠信息科技有限公司 | Interactive icon display method, device, medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102444777B1 (en) | Routing messages by message parameter | |
| US20150112753A1 (en) | Social content filter to enhance sentiment analysis | |
| CN108369709B (en) | System and method for network-based advertisement data traffic latency reduction | |
| CN106874314B (en) | Information recommendation method and device | |
| US10268897B2 (en) | Determining most representative still image of a video for specific user | |
| US10402479B2 (en) | Method, server, browser, and system for recommending text information | |
| CN109446378A (en) | Information recommendation method, Sentiment orientation determine method and device and electronic equipment | |
| US10387915B2 (en) | Digital magazine recommendations by topic | |
| US20110093520A1 (en) | Automatically identifying and summarizing content published by key influencers | |
| WO2017112017A1 (en) | Interactive commentary based on video-content characteristics | |
| WO2017019460A1 (en) | Automatic artist and content breakout prediction | |
| WO2014194689A1 (en) | Method, server, browser, and system for recommending text information | |
| CN112395489A (en) | Recommendation method, recommendation device, recommendation equipment and computer storage medium | |
| US9575996B2 (en) | Emotion image recommendation system and method thereof | |
| US20150262238A1 (en) | Techniques for Topic Extraction Using Targeted Message Characteristics | |
| CN107943789A (en) | Mood analysis method, device and the server of topic information | |
| CN107578263A (en) | A kind of detection method, device and the electronic equipment of advertisement abnormal access | |
| CN105302880A (en) | Content correlation recommendation method and apparatus | |
| CN105812353A (en) | Terminal identification method and apparatus | |
| CN111159572A (en) | Recommended content auditing method and device, electronic equipment and storage medium | |
| US20110295780A1 (en) | Apparatus and Method for Personalized Delivery of Content from Multiple Data Sources | |
| CN108401005B (en) | Expression recommendation method and device | |
| US20180032907A1 (en) | Detecting abusive language using character n-gram features | |
| CN106993226A (en) | A kind of method and terminal of recommendation video | |
| US20130159271A1 (en) | Graphical user interface for web browser |
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: 20190308 |
|
| RJ01 | Rejection of invention patent application after publication |