CN110245274A - A kind of label temperature calculates method, apparatus, electronic equipment and storage medium - Google Patents
A kind of label temperature calculates method, apparatus, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110245274A CN110245274A CN201910329458.4A CN201910329458A CN110245274A CN 110245274 A CN110245274 A CN 110245274A CN 201910329458 A CN201910329458 A CN 201910329458A CN 110245274 A CN110245274 A CN 110245274A
- Authority
- CN
- China
- Prior art keywords
- behavior
- user
- label
- hot value
- temperature
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/954—Navigation, e.g. using categorised browsing
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of label temperatures to calculate method, apparatus, electronic equipment and storage medium, and according to the behavioural information of each user, the user behavior to match with behavior mark is determined goal behavior;It according to the initial weighting coefficients and behavior time of origin of the attenuation coefficient of user tag and each goal behavior, is weighted, obtains the corresponding total label temperature residual volume of current time;Label hot value according to total label temperature residual volume and section of next die-away time, after calculating decaying.This method can embody influence of each user to the label temperature of the same user tag, and in the final label hot value of determination, consider over time, label temperature generates the case where decaying, and the label hot value after the decaying calculated is enabled more precisely to embody the corresponding label hot value of current time.As it can be seen that this method can more precisely reflect the variable quantity of user behavior preference and user behavior preference.
Description
Technical field
The present invention relates to Internet technical field more particularly to a kind of label temperature calculate method, apparatus, electronic equipment and
Storage medium.
Background technique
To allow user to find interested content in the content of magnanimity, APP can believe according to the historical viewings of user
It ceases to the user and pushes relevant information.
And realize user push relevant information when, APP can establish in the form of a label user behavior and browsing information it
Between incidence relation, in the form of different user label characterize user for different information preference, with different label Thermometers
Family is taken over for use for the fancy grade of different information.For example, if user A and user B repeatedly browse the second-hand house in two Room, one Room
Information can establish " 2 bedrooms " and " room " the two user tags for user A and user B;If in same time
In section, user A browses the frequency and is higher than user B, then the label temperature of " 2 bedrooms " of user A and " room " is higher than user B
" 2 bedrooms " and " room " label temperature.
Applicant has found that over time, user may be continuous to the fancy grade of a certain information under study for action
Decaying, the corresponding label temperature of corresponding user tag also should gradually decay, and therefore, for energy, accurate recommended user is interested
Information, also need to be adjusted the corresponding label temperature of user tag, so according to adjustment label temperature after user tag
For user's pushed information.
The common method of adjustment label temperature is that the forgetting of user is fitted by Newtonian Cooling method using Newtonian Cooling method
Rule, and then deduce out the label temperature variation of each user tag.But Newtonian Cooling method is only applicable to calculate sole user
Influence of the behavior for label temperature, and user can generate a variety of user behaviors when operating APP, different user behaviors also can
Different liveness is generated, so that the calculation method can not really reflect user for the degree of information preference.
Summary of the invention
The present invention provides a kind of label temperatures to calculate method, apparatus, electronic equipment and storage medium, existing to solve
The low problem of label temperature calculation method accuracy rate.
In a first aspect, the present invention provides a kind of label temperature calculation methods, comprising the following steps:
Obtain the temperature attenuation coefficient and the corresponding user behavior information of the user tag of user tag, user's row
It include user behavior, behavior mark, behavior time of origin and initial weighting coefficients for information;
It is identified according to the behavior and determines goal behavior, the goal behavior, which refers to, identifies the user's row to match with behavior
For;
According to initial weighting coefficients corresponding to the attenuation coefficient of the user tag and each goal behavior and
Behavior time of origin, is weighted, and obtains the corresponding total label temperature residual volume of current time;
Label hot value according to total label temperature residual volume and section of next die-away time, after calculating decaying.
Further,
Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is initial weighting coefficients, and k is user tag
Attenuation coefficient, △ t are the time difference of behavior time of origin and current time.
Further,
T (t+n)=R × e-kn;
Wherein, T is the label hot value after decaying, and R is total label temperature residual volume, and k is the attenuation coefficient of label, and n is
Next die-away time section, t is current time.
Further, further includes:
According to the label hot value after the decaying, the hot value that each user generates the user tag is determined;
The hot value that each user to user label generates is ranked up, divides temperature according to the ranking results
Grade;
According to the sequence of the Heat range from high to low, the new label hot value and new label temperature of corresponding user are set
Grade.
Further, the goal behavior includes click behavior, splitting glass opaque, collection behavior, short message behavior, system chat
Behavior and phone behavior.
Second aspect, the present invention also provides a kind of label temperature computing devices, comprising:
Data obtaining module, the corresponding user's row of temperature attenuation coefficient and the user tag for obtaining user tag
For information, the user behavior information includes user behavior, behavior mark, behavior time of origin and initial weighting coefficients;
Goal behavior determining module, for according to the behavior identify determine goal behavior, the goal behavior refer to
Behavior identifies the user behavior to match;
Weighted calculation module, for right according to the attenuation coefficient of the user tag and each goal behavior institute
The initial weighting coefficients and behavior time of origin answered, are weighted, and it is remaining to obtain the corresponding total label temperature of current time
Amount;
Label hot value computing module, for calculating according to total label temperature residual volume and section of next die-away time
Label hot value after decaying.
Further,
Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is initial weighting coefficients, and k is user tag
Attenuation coefficient, △ t are the time difference of behavior time of origin and current time.
Further,
T (t+n)=R × e-kn;
Wherein, T is the label hot value after decaying, and R is total label temperature residual volume, and k is the attenuation coefficient of label, and n is
Next die-away time section, t is current time.
Further, further includes:
Hot value determining module, for determining each user to the user according to the label hot value after the decaying
The hot value that label generates;
Grade classification module, the hot value for generating each user to user label is ranked up, according to institute
It states ranking results and divides Heat range;
Setting module sets the new label temperature of corresponding user for the sequence according to the Heat range from high to low
Value and new label Heat range.
Further, the goal behavior includes click behavior, splitting glass opaque, collection behavior, short message behavior, system chat
Behavior and phone behavior.
The third aspect, the present invention also provides a kind of electronic equipment, comprising: memory, for storing program instruction;
Processor, for calling and executing the program instruction in the memory, to realize label described in first aspect
Temperature calculation method.
Fourth aspect is stored with computer program in the storage medium the present invention also provides a kind of storage medium, when
When at least one processor of label temperature computing device executes the computer program, label temperature computing device executes first
Label temperature calculation method described in aspect.
From the above technical scheme, the embodiment of the invention provides a kind of label temperatures to calculate method, apparatus, electronics is set
The user behavior to match with behavior mark is determined goal behavior according to the behavioural information of each user by standby and storage medium;
According to the initial weighting coefficients and behavior time of origin of the attenuation coefficient of user tag and each goal behavior, it is weighted
It calculates, obtains the corresponding total label temperature residual volume of current time;According to total label temperature residual volume and next die-away time section,
Label hot value after calculating decaying.This method can embody each user to the shadow of the label temperature of the same user tag
It rings, and in the final label hot value of determination, considers that over time, label temperature generates the case where decaying, so that
Label hot value after the decaying of calculating can more precisely embody the corresponding label hot value of current time.As it can be seen that the party
Method can more precisely reflect the variable quantity of user behavior preference and user behavior preference.
Detailed description of the invention
In order to illustrate more clearly of technical solution of the present invention, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow chart of label temperature calculation method provided in an embodiment of the present invention;
Fig. 2 is another flow chart of label temperature calculation method provided in an embodiment of the present invention;
Fig. 3 is the structural block diagram of label temperature computing device provided in an embodiment of the present invention;
Fig. 4 is the hardware structural diagram of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Fig. 1 is the flow chart of label temperature calculation method provided in an embodiment of the present invention.
A kind of label temperature calculation method provided in an embodiment of the present invention, improves existing Newtonian Cooling method, obtains
To the Newtonian Cooling method of Weighted Coefficients, accurate calculation is carried out to the temperature that multiple user behaviors generate using this method, and utilize heat
Attenuation model is spent, accurately determines every kind of user tag in the remaining label temperature after attenuation process, more realistically to react
User behavior preference and the variation of user behavior preference.
Specifically, referring to Fig. 1, a kind of label temperature calculation method provided in an embodiment of the present invention, comprising the following steps:
S1, the temperature attenuation coefficient and the corresponding user behavior information of user tag for obtaining user tag, user behavior letter
Breath includes user behavior, behavior mark, behavior time of origin and initial weighting coefficients.
Over time, the present embodiment is when determining the corresponding label temperature variation of each user tag, foundation
It is the behavioural information of user, and, the corresponding temperature attenuation coefficient of each user tag.
User behavior information is used to record movement when each user's operation APP, including user behavior, behavior mark, row
For time of origin and initial weighting coefficients.Operation behavior that user behavior includes user in the version feed stream of local, its in APP
Behavior etc. in his behavior, individual subscriber center.Behavior of the user inside the version feed stream of local, including user for model
Click behavior, system chat behavior, behavior of making a phone call, behavior of sending short messages, splitting glass opaque and collection behavior etc..Other in APP
Behavior includes sliding behavior, selection service item behavior etc..Behavior in individual subscriber center include setting personal preference behavior,
Mask information behavior etc..
Behavior mark is indicated for characterizing user behavior, each user behavior by a behavior mark.Behavior mark
It is stored in User action log, when User action log to be sent in system, behavior mark can be carried and sent together, used
In judging whether user behavior is legal.Behavior time of origin refers to time of the user when first time generating corelation behaviour.Initially
Weighting coefficient is the corresponding weighting coefficient of every kind of user behavior, is generated not for characterizing different user to different user tags
Same liveness, to determine influence of each user to label temperature.
Label temperature reflects user for the preference of such model, and label temperature is higher, and preference is higher.It is different
User behavior reflection user preference it is also different, such as: user clicks one second-hand house model and user on APP leads to
Cross phone contact this second-hand house publisher.Both behaviors can all reflect user for the preference of second-hand house category, but it is obvious
The preference of second situation user is stronger.Comprehensive temperature be by user on APP all lawful acts all carry out statistics and
It calculates, calculates the preference of a totality.
Temperature attenuation coefficient is used to be characterized in the variable quantity of different moments temperature.Because can produce the user of the user tag
Behavior has very much, and different user behaviors can also generate different liveness.And over time, user continues with one
Certain customers' behavior browses the relevant information of the user tag, in addition, perhaps user does not recycle certain a user behavior browsings
The relevant information of the user tag causes for different user behavior, and the corresponding label temperature of same user tag can be different, because
This, the temperature attenuation coefficient of every kind of user tag can be different.
The present embodiment calculates the remaining label temperature of user tag according to temperature attenuation coefficient, and calculated result can be improved
Accuracy.
S2, determining goal behavior is identified according to behavior, goal behavior, which refers to, identifies the user behavior to match with behavior.
Behavior is identified for judging whether active user's behavior is legal, is judging active user's behavior according to behavior mark
When legal, which is determined as goal behavior.Goal behavior is the use having a direct impact to the variation of label temperature
Family behavior, and goal behavior need to identify with behavior and match, otherwise the user behavior is judged as illegal.
Specifically, in the present embodiment, it can include click behavior to the goal behavior that label temperature has a direct impact, share
Behavior, collection behavior, short message behavior, system chat behavior and phone behavior.Click behavior is that user clicks some model to look into
See the behavior of details;Splitting glass opaque is the behavior that some model is shared with another user by user;Collection behavior can body
Reveal user to the degree deeply concerned of the information, convenient for checking again in the future;Short message behavior is that user exchanges with businessman's progress short message
Behavior;System chat behavior is the behavior that user is exchanged with businessman by the chat program that APP is carried;Phone behavior is
The behavior that user is directly exchanged by dialing the phone in information with businessman.
Different behaviors illustrate the user to the different interest levels of information, therefore, by it is above-mentioned can be directly to label
The goal behavior that temperature has an impact calculates the remaining label temperature of user tag, and the accuracy of calculated result can be improved.
Initial weighting coefficients and behavior corresponding to S3, attenuation coefficient and each goal behavior according to user tag
Time of origin is weighted, and obtains the corresponding total label temperature residual volume of current time.
Current time is to calculate the current time of label temperature, since label temperature constantly decays over time,
Therefore, it is the accurate surplus for determining label temperature, needs first to determine total mark that each user behavior generates the user tag
Sign temperature residual volume.
Specifically, the present embodiment determines the corresponding total label temperature residual volume of current time according to the following formula:
Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is initial weighting coefficients, and k is user tag
Attenuation coefficient, △ t are the time difference of behavior time of origin and current time.
Since the different corresponding initial weighting coefficients of user behavior are different, in the present embodiment, click the initial of behavior and add
Weight coefficient is 1, and the initial weighting coefficients of splitting glass opaque are 2, and the initial weighting coefficients for collecting behavior are 2, short message behavior it is initial
Weighting coefficient is 4, and the initial weighting coefficients of system chat behavior are 4, and the initial weighting coefficients of phone behavior are 5.Every kind of user
The initial weighting coefficients of behavior can be also other, and the present embodiment is not specifically limited.
Different user tags corresponds to different attenuation coefficients, therefore, when calculating the label temperature of same user tag,
Attenuation coefficient may be configured as 1.△ t is the time difference of behavior time of origin and current time, calculates the label heat in this time
Degree variation can specify total residual volume of the corresponding label temperature of current time.
When calculating label temperature using existing Newtonian Cooling method, need to calculate separately the label of every kind of user behavior generation
After temperature, then label temperature summation that each user behavior is generated, the label temperature of the user tag is determined.The process meter
Calculation amount is larger, and therefore, the present embodiment improves on the basis of existing Newtonian Cooling method, calculates by deducing and simplifying
Journey obtains the Newtonian Cooling method of Weighted Coefficients, and influence of a variety of user behaviors for label temperature is handled by this method, calculates
Label temperature, can be improved computational efficiency, also avoids individually calculating the situation for mistake easily occur, improves and calculate accuracy.
S4, the label hot value according to total label temperature residual volume and section of next die-away time, after calculating decaying.
As time goes by, after determining current time corresponding total label temperature residual volume, using one
Section time, i.e. next die-away time section determines the label hot value after decaying with this.
In the present embodiment, according to the following formula, the label hot value after determining decaying:
T (t+n)=R × e-kn。
Wherein, T is the label hot value after decaying, and R is total label temperature residual volume, and k is the attenuation coefficient of label, and n is
Next die-away time section, t is current time.
Using the Newtonian Cooling method of Weighted Coefficients, i.e. formula:After next die-away time section n,
Label hot value after determining decaying, label hot value and total label using the deduction of some column, that is, after can determine decaying
Relationship between temperature residual volume is as follows:
As it can be seen that by deducing, the attenuation results and each specific mesh of the corresponding label temperature of available user tag
Mark behavior is unrelated, and only related with next die-away time section and initial weighting coefficients, thus can accurately determine that user tag exists
After a period of time, the label hot value after decaying.
In addition, above-described embodiment be using next die-away time section total label temperature residual volume corresponding with current time come
The final label hot value determined after decaying also can be used in another embodiment and utilize next die-away time section and upper one
The method of temperature time corresponding total label temperature residual volume is calculated to determine label hot value after the corresponding decaying of present period.
Two methods are only that time conditions when choosing temperature residual volume are different, but the Computing Principle of the two be it is identical, therefore, should
Kind method is also the protection scope of the present embodiment.
Every kind of user tag is being determined after undergoing the label hot value after overdamping, system should be according to current label
Hot value is user's pushed information, therefore, it is necessary to carry out ranking to temperature according to the current label hot value of user tag, with more
Add accurately is user according to current preference pushed information.
Specifically, label temperature calculation method provided in this embodiment, as shown in Figure 2, further includes:
S5, according to the label hot value after decaying, determine the hot value that each user to user label generates.
Label hot value after decaying is the remaining hot value that the different behaviors of multiple users generate the user tag, is
Accomplish the accurate pushed information for each user, to be pushed according to the current preference of relative users, needs again true
The current temperature that fixed each user generates the user tag.
S6, the hot value that each user to user label generates is ranked up, divides Heat range according to ranking results.
All users for possessing this user tag are subjected to ranking according to temperature size, and according to rank order division etc.
Grade.Sequence i.e. descending according to temperature, user is ranked up, and determines rank order point, for example, determination ranks preceding 20%
The rank order of user be divided into 5 points, determine that the rank order for the user for ranking 20%~40% is divided into 4 points, determination is ranked
The rank order of 40%~60% user is divided into 3 points, determines that the rank order for the user for ranking 60%~80% is divided into 2 points,
It determines that the rank order for the user for ranking 80%~100% is divided into 1 point, divides Heat range according to rank order graduation.
S7, the sequence according to Heat range from high to low set the new label hot value and new label temperature of corresponding user
Grade.
The higher corresponding Heat range of rank order point is also higher, the lower corresponding Heat range of rank order point
It is lower, and so on.It is the higher corresponding user of Heat range, Hold sticker hot value, and Hold sticker heat in the present embodiment
Degree grade, and the relatively low corresponding user of remaining Heat range, then redefine its new label hot value and new label temperature
Grade.
The present embodiment characterizes the influence that user generates label temperature, again by dividing Heat range really to reflect
Preference of the user for model out.
From the above technical scheme, the embodiment of the invention provides a kind of label temperature calculation methods, according to each use
The user behavior to match with behavior mark is determined goal behavior by the behavioural information at family;According to the attenuation coefficient of user tag,
And the initial weighting coefficients and behavior time of origin of each goal behavior, it is weighted, it is corresponding to obtain current time
Total label temperature residual volume;Label hot value according to total label temperature residual volume and section of next die-away time, after calculating decaying.
This method can embody influence of each user to the label temperature of the same user tag, and the label heat final in determination
When angle value, consider that over time, label temperature generates the case where decaying, so that the label hot value after the decaying calculated
The corresponding label hot value of current time can more precisely be embodied.As it can be seen that this method can more precisely reflect user
The variable quantity of Behavior preference and user behavior preference.
As shown in figure 3, the present invention also provides a kind of label temperature computing device, for executing label heat shown in FIG. 1
The correlation step of calculation method is spent, which includes:
Data obtaining module 10, for obtaining temperature attenuation coefficient and the corresponding user of the user tag of user tag
Behavioural information, the user behavior information include user behavior, behavior mark, behavior time of origin and initial weighting coefficients;Mesh
Mark behavior determining module 20 determines goal behavior for identifying according to the behavior, and the goal behavior refers to be identified with behavior
The user behavior to match;Weighted calculation module 30, for the attenuation coefficient and each mesh according to the user tag
Initial weighting coefficients corresponding to mark behavior and behavior time of origin, are weighted, and obtain the corresponding total mark of current time
Sign temperature residual volume;Label hot value computing module 40, for according to total label temperature residual volume and next die-away time
Section, the label hot value after calculating decaying.
Further,Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is first
Beginning weighting coefficient, k are the attenuation coefficient of user tag, and △ t is the time difference of behavior time of origin and current time.
Further, T (t+n)=R × e-kn;Wherein, T is the label hot value after decaying, and R is that total label temperature is remaining
Amount, k are the attenuation coefficient of label, and n is next die-away time section, and t is current time.
Further, further includes: hot value determining module, for determining every according to the label hot value after the decaying
The hot value that a user generates the user tag;Grade classification module, for producing each user to user label
Raw hot value is ranked up, and divides Heat range according to the ranking results;Setting module, for according to the Heat range
Sequence from high to low sets the new label hot value for corresponding to user and new label Heat range.
Further, the goal behavior includes click behavior, splitting glass opaque, collection behavior, short message behavior, system chat
Behavior and phone behavior.
Fig. 4 is the hardware structural diagram of another electronic equipment provided in an embodiment of the present invention.As shown in figure 4, this hair
It is bright to additionally provide a kind of electronic equipment, comprising: memory 601, for storing program instruction;Processor 602, for calling and holding
Program instruction in the row memory, to realize label temperature calculation method described in above-described embodiment.Specifically it may refer to
Associated description in previous embodiment.
In the present embodiment, processor 602 can be connected with memory 601 by bus or other modes.Processor can be
General processor, such as central processing unit, digital signal processor, specific integrated circuit, or be configured to implement the present invention
One or more integrated circuits of embodiment.Memory may include volatile memory, such as random access memory;Storage
Device also may include nonvolatile memory, such as read-only memory, flash memory, hard disk or solid state hard disk.
The embodiment of the invention also provides a kind of storage medium, it is stored with computer program in the storage medium, works as mark
When signing at least one processor execution computer program of temperature computing device, label temperature computing device executes above-mentioned reality
Apply label temperature calculation method described in example.
The storage medium can be magnetic disk, CD, read-only memory (English: read-only memory, letter
Claim: ROM) or random access memory (English: random access memory, referred to as: RAM) etc..
It is required that those skilled in the art can be understood that the technology in the embodiment of the present invention can add by software
The mode of general hardware platform realize.Based on this understanding, the technical solution in the embodiment of the present invention substantially or
Say that the part that contributes to existing technology can be embodied in the form of software products, which can deposit
Storage is in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) execute certain part institutes of each embodiment of the present invention or embodiment
The method stated.
Same and similar part may refer to each other between each embodiment in this specification.Especially for label temperature
For computing device embodiment, since it is substantially similar to the method embodiment, so be described relatively simple, related place referring to
Explanation in embodiment of the method.
Invention described above embodiment is not intended to limit the scope of the present invention..
Claims (12)
1. a kind of label temperature calculation method, which comprises the following steps:
Obtain the temperature attenuation coefficient and the corresponding user behavior information of the user tag of user tag, the user behavior letter
Breath includes user behavior, behavior mark, behavior time of origin and initial weighting coefficients;
It is identified according to the behavior and determines goal behavior, the goal behavior, which refers to, identifies the user behavior to match with behavior;
According to initial weighting coefficients and behavior corresponding to the attenuation coefficient of the user tag and each goal behavior
Time of origin is weighted, and obtains the corresponding total label temperature residual volume of current time;
Label hot value according to total label temperature residual volume and section of next die-away time, after calculating decaying.
2. the method according to claim 1, wherein
Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is initial weighting coefficients, and k is the decaying of user tag
Coefficient, △ t are the time difference of behavior time of origin and current time.
3. the method according to claim 1, wherein
T (t+n)=R × e-kn;
Wherein, T is the label hot value after decaying, and R is total label temperature residual volume, and k is the attenuation coefficient of label, and n is next
Die-away time section, t is current time.
4. the method according to claim 1, wherein further include:
According to the label hot value after the decaying, the hot value that each user generates the user tag is determined;
The hot value that each user to user label generates is ranked up, divides temperature etc. according to the ranking results
Grade;
According to the sequence of the Heat range from high to low, the new label hot value for corresponding to user and new label temperature etc. are set
Grade.
5. the method according to claim 1, wherein the goal behavior includes click behavior, splitting glass opaque, receipts
Hiding behavior, short message behavior, system chat behavior and phone behavior.
6. a kind of label temperature computing device characterized by comprising
Data obtaining module, the corresponding user behavior letter of temperature attenuation coefficient and the user tag for obtaining user tag
Breath, the user behavior information includes user behavior, behavior mark, behavior time of origin and initial weighting coefficients;
Goal behavior determining module determines goal behavior for identifying according to the behavior, and the goal behavior refers to and behavior
Identify the user behavior to match;
Weighted calculation module, for corresponding to the attenuation coefficient and each goal behavior according to the user tag
Initial weighting coefficients and behavior time of origin, are weighted, and obtain the corresponding total label temperature residual volume of current time;
Label hot value computing module, for calculating decaying according to total label temperature residual volume and section of next die-away time
Label hot value afterwards.
7. device according to claim 6, which is characterized in that
Wherein, R is total label temperature residual volume, and N is N kind goal behavior, and r is initial weighting coefficients, and k is the decaying of user tag
Coefficient, △ t are the time difference of behavior time of origin and current time.
8. device according to claim 6, which is characterized in that
T (t+n)=R × e-kn;
Wherein, T is the label hot value after decaying, and R is total label temperature residual volume, and k is the attenuation coefficient of label, and n is next
Die-away time section, t is current time.
9. device according to claim 6, which is characterized in that further include:
Hot value determining module, for determining each user to the user tag according to the label hot value after the decaying
The hot value of generation;
Grade classification module, the hot value for generating each user to user label is ranked up, according to the row
Sequence result divides Heat range;
Setting module, for the sequence according to the Heat range from high to low, set corresponding user new label hot value and
New label Heat range.
10. device according to claim 6, which is characterized in that the goal behavior include click behavior, splitting glass opaque,
Collection behavior, short message behavior, system chat behavior and phone behavior.
11. a kind of electronic equipment characterized by comprising
Memory, for storing program instruction;
Processor, for calling and executing the program instruction in the memory, to realize described in any one of Claims 1 to 5
Label temperature calculation method.
12. a kind of storage medium, which is characterized in that computer program is stored in the storage medium, when label temperature calculates
When at least one processor of device executes the computer program, the requirement 1~5 of label temperature computing device perform claim is any
Label temperature calculation method described in.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910329458.4A CN110245274A (en) | 2019-04-23 | 2019-04-23 | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910329458.4A CN110245274A (en) | 2019-04-23 | 2019-04-23 | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110245274A true CN110245274A (en) | 2019-09-17 |
Family
ID=67883236
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910329458.4A Pending CN110245274A (en) | 2019-04-23 | 2019-04-23 | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110245274A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110599052A (en) * | 2019-09-19 | 2019-12-20 | 携程计算机技术(上海)有限公司 | OTA hotel evaluation method, system, electronic device and medium |
CN110719506A (en) * | 2019-10-21 | 2020-01-21 | 广州酷狗计算机科技有限公司 | Method, device, server and storage medium for determining interest degree of user in video |
CN111178950A (en) * | 2019-12-19 | 2020-05-19 | 车智互联(北京)科技有限公司 | User portrait construction method and device and computing equipment |
CN111415328A (en) * | 2020-02-20 | 2020-07-14 | 北京三快在线科技有限公司 | Article analysis data determination method and device and electronic equipment |
CN111581452A (en) * | 2020-03-26 | 2020-08-25 | 浙江口碑网络技术有限公司 | Method and device for obtaining recommendation object data and electronic equipment |
CN112632389A (en) * | 2020-12-30 | 2021-04-09 | 广州博冠信息科技有限公司 | Information processing method, information processing apparatus, storage medium, and electronic device |
CN112883267A (en) * | 2021-02-22 | 2021-06-01 | 深圳市星网储区块链有限公司 | Data heat degree statistical method and device based on deep learning |
CN113806688A (en) * | 2021-09-30 | 2021-12-17 | 中孚安全技术有限公司 | Label influence degree calculation method and system based on exponential decay |
CN114968029A (en) * | 2022-06-09 | 2022-08-30 | 合众新能源汽车有限公司 | Popular menu presenting method and device based on crowdsourcing and computer readable medium |
CN114968030A (en) * | 2022-06-09 | 2022-08-30 | 合众新能源汽车有限公司 | Presentation method, presentation device and computer readable medium of self-adaptive personalized menu |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108304399A (en) * | 2017-01-12 | 2018-07-20 | 武汉斗鱼网络科技有限公司 | The recommendation method and device of Web content |
CN109040844A (en) * | 2018-09-25 | 2018-12-18 | 有米科技股份有限公司 | A kind of method, apparatus and electronic equipment obtaining video temperature |
CN109657138A (en) * | 2018-12-10 | 2019-04-19 | 深圳墨世科技有限公司 | A kind of video recommendation method, device, electronic equipment and storage medium |
-
2019
- 2019-04-23 CN CN201910329458.4A patent/CN110245274A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108304399A (en) * | 2017-01-12 | 2018-07-20 | 武汉斗鱼网络科技有限公司 | The recommendation method and device of Web content |
CN109040844A (en) * | 2018-09-25 | 2018-12-18 | 有米科技股份有限公司 | A kind of method, apparatus and electronic equipment obtaining video temperature |
CN109657138A (en) * | 2018-12-10 | 2019-04-19 | 深圳墨世科技有限公司 | A kind of video recommendation method, device, electronic equipment and storage medium |
Non-Patent Citations (1)
Title |
---|
LITTLEMAGIC: "设计一个属于自己的热度值算法", 《HTTPS://WWW.JIANSHU.COM/P/FB454A4B383D》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110599052A (en) * | 2019-09-19 | 2019-12-20 | 携程计算机技术(上海)有限公司 | OTA hotel evaluation method, system, electronic device and medium |
CN110599052B (en) * | 2019-09-19 | 2023-07-21 | 携程计算机技术(上海)有限公司 | OTA hotel evaluation method, system, electronic equipment and medium |
CN110719506B (en) * | 2019-10-21 | 2022-02-11 | 广州酷狗计算机科技有限公司 | Method, device, server and storage medium for determining interest degree of user in video |
CN110719506A (en) * | 2019-10-21 | 2020-01-21 | 广州酷狗计算机科技有限公司 | Method, device, server and storage medium for determining interest degree of user in video |
CN111178950A (en) * | 2019-12-19 | 2020-05-19 | 车智互联(北京)科技有限公司 | User portrait construction method and device and computing equipment |
CN111178950B (en) * | 2019-12-19 | 2023-08-29 | 车智互联(北京)科技有限公司 | User portrait construction method and device and computing equipment |
CN111415328A (en) * | 2020-02-20 | 2020-07-14 | 北京三快在线科技有限公司 | Article analysis data determination method and device and electronic equipment |
CN111415328B (en) * | 2020-02-20 | 2023-07-11 | 北京三快在线科技有限公司 | Method and device for determining article analysis data and electronic equipment |
CN111581452A (en) * | 2020-03-26 | 2020-08-25 | 浙江口碑网络技术有限公司 | Method and device for obtaining recommendation object data and electronic equipment |
CN111581452B (en) * | 2020-03-26 | 2023-10-17 | 浙江口碑网络技术有限公司 | Recommendation object data obtaining method and device and electronic equipment |
CN112632389A (en) * | 2020-12-30 | 2021-04-09 | 广州博冠信息科技有限公司 | Information processing method, information processing apparatus, storage medium, and electronic device |
CN112632389B (en) * | 2020-12-30 | 2024-03-15 | 广州博冠信息科技有限公司 | Information processing method, information processing apparatus, storage medium, and electronic device |
CN112883267A (en) * | 2021-02-22 | 2021-06-01 | 深圳市星网储区块链有限公司 | Data heat degree statistical method and device based on deep learning |
CN113806688A (en) * | 2021-09-30 | 2021-12-17 | 中孚安全技术有限公司 | Label influence degree calculation method and system based on exponential decay |
CN114968030A (en) * | 2022-06-09 | 2022-08-30 | 合众新能源汽车有限公司 | Presentation method, presentation device and computer readable medium of self-adaptive personalized menu |
CN114968029A (en) * | 2022-06-09 | 2022-08-30 | 合众新能源汽车有限公司 | Popular menu presenting method and device based on crowdsourcing and computer readable medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110245274A (en) | A kind of label temperature calculates method, apparatus, electronic equipment and storage medium | |
CN109634819B (en) | Alarm root cause positioning method and device and electronic equipment | |
US20200356572A1 (en) | Search ranking method and apparatus, electronic device and storage medium | |
CN107402881A (en) | The choosing method and device of a kind of project testing | |
CN109309596B (en) | Pressure testing method and device and server | |
CN110378474A (en) | Fight sample generating method, device, electronic equipment and computer-readable medium | |
CN108363730B (en) | Content recommendation method, system and terminal equipment | |
CN109815416A (en) | Method for pushing, the device of related information, electronic equipment and storage medium | |
CN107273508A (en) | Information processing method and device based on artificial intelligence | |
CN108965951B (en) | Advertisement playing method and device | |
CN106257448A (en) | The methods of exhibiting of a kind of key word and device | |
CN105871695B (en) | Expression sending method and device | |
CN110046944A (en) | Invoice creation method and device, electronic equipment based on block chain | |
CN113076416A (en) | Information heat evaluation method and device and electronic equipment | |
CN107766106B (en) | The method and apparatus for generating configuration file | |
CN109919656A (en) | Advertisement placement method, device, computer equipment and storage medium | |
CN113422986A (en) | Method, apparatus, device, medium, and program product for live room recommendation | |
CN105677645B (en) | A kind of tables of data comparison method and device | |
CN106156351B (en) | Multimedia resource recommendation information generation method and device | |
CN106452808A (en) | Data processing method and data processing device | |
JP2019503014A (en) | Method and apparatus for processing user behavior data | |
CN113449188A (en) | Application recommendation method and device, electronic equipment and readable storage medium | |
CN109325227A (en) | Method and apparatus for generating amendment sentence | |
CN108228598B (en) | Media information sorting method, server and system | |
CN111415197A (en) | Method for acquiring article exchange voucher, method for providing article exchange voucher, device and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |