CN109447462A - It is a kind of for determining the method and apparatus of the impact factor of research object - Google Patents
It is a kind of for determining the method and apparatus of the impact factor of research object Download PDFInfo
- Publication number
- CN109447462A CN109447462A CN201811265351.XA CN201811265351A CN109447462A CN 109447462 A CN109447462 A CN 109447462A CN 201811265351 A CN201811265351 A CN 201811265351A CN 109447462 A CN109447462 A CN 109447462A
- Authority
- CN
- China
- Prior art keywords
- information
- impact factor
- research object
- vector
- evaluation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The purpose of the application is to provide a kind of method and apparatus for determining the impact factor of research object, determines the cognitive appraisal information of the research object, pays close attention to evaluation information and enliven evaluation information；Based on the cognitive appraisal information, the concern evaluation information and it is described enliven evaluation information, determine corresponding cognitive appraisal vector respectively, pay close attention to evaluation vector and enliven evaluation vector；The cognitive appraisal vector, the concern evaluation vector and the evaluation vector that enlivens are determined into the current impact factor of the research object applied to the impact factor model about the research object, and based on the output of the impact factor model.The application, which achieves, to save overhead and guarantees the balance between system feedback timeliness, and can obtain accurate impact factor.
Description
Technical field
This application involves computer fields more particularly to a kind of for determining the technology of the impact factor of research object.
Background technique
With the development of information technology, masses be able to understand a variety of research objects (for example including but be not limited to each type games
The public figures such as member, performer or all kinds of above-the-line projects) and the activities such as comment on, give a mark, and these activities are to research pair
The correlated results as caused by (or impact factor) causes increasing influence to the influence power of all kinds of research objects, public
Also above-mentioned impact factor is relied on more and more and as relevant action (such as participate in topic, make relevant decision etc.)
Foundation.The impact factor data an of research object are obtained in time and accurately, it will help masses make correct decisions.
Summary of the invention
The purpose of the application is to provide a kind of method for determining the impact factor of research object.
According to the one aspect of the application, a kind of method for determining the impact factor of research object, the party are provided
Method the following steps are included:
It determines the cognitive appraisal information of the research object, pay close attention to evaluation information and enliven evaluation information；
Based on the cognitive appraisal information, the concern evaluation information and it is described enliven evaluation information, respectively determine pair
The cognitive appraisal vector answered pays close attention to evaluation vector and enlivens evaluation vector；
The cognitive appraisal vector, the concern evaluation vector and the evaluation vector that enlivens are applied to about described
The impact factor model of research object, and the current influence of the research object is determined based on the output of the impact factor model
The factor.
According to further aspect of the application, a kind of equipment for determining the impact factor of research object is provided, it should
Equipment includes:
First module, for determining cognitive appraisal information, concern evaluation information and the active evaluation of the research object
Information；
Second module, for based on the cognitive appraisal information, the concern evaluation information and the active evaluation letter
Breath, determines corresponding cognitive appraisal vector respectively, pays close attention to evaluation vector and enliven evaluation vector；
Third module, for by the cognitive appraisal vector, the concern evaluation vector and described enlivening evaluation vector
The research is determined applied to the impact factor model about the research object, and based on the output of the impact factor model
The current impact factor of object.
According to further aspect of the application, a kind of equipment for determining the impact factor of research object is provided, it should
Equipment includes:
Processor；And
It is arranged to the memory of storage computer executable instructions, the executable instruction makes the place when executed
It manages device and executes the process described above.
According to further aspect of the application, a kind of computer-readable medium including instruction is provided, described instruction exists
It is performed so that system executes the process described above.
Compared with prior art, the application is based on context analyzer, temperature analysis and the mood analysis difference to research object
The cognitive appraisal information of research object is obtained, evaluation information is paid close attention to and enlivens evaluation information, and based on these information and relevant
Data model obtains the impact factor of research object, has fully considered the slower cognitive appraisal information of opposite variation and opposite variation
Faster concern evaluation information saves overhead and guarantees the balance between system feedback timeliness to achieve, and
Accurate impact factor can be obtained；Further, after the backtracking data for introducing impact factor, the application can also be protected further
It holds the stability of the impact factor of research object and keeps its relatively reliable.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is according to a kind of for determining the process of the method for the impact factor of research object of the application one embodiment
Figure；
Fig. 2 shows according to a kind of for determining the equipment of the impact factor of research object of the application another embodiment
Functional module；
Fig. 3 shows a kind of exemplary system according to another embodiment of the application.
The same or similar appended drawing reference represents the same or similar component in attached drawing.
Specific embodiment
The application is described in further detail with reference to the accompanying drawing.
In a typical configuration of this application, terminal, the equipment of service network and trusted party include one or more
Processor (CPU), input/output interface, network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or
Any other non-transmission medium, can be used for storage can be accessed by a computing device information.
The application meaning equipment includes but is not limited to that user equipment, the network equipment or user equipment and the network equipment pass through
Network is integrated constituted equipment.The user equipment includes but is not limited to that any one can carry out human-computer interaction with user
The mobile electronic product, such as smart phone, tablet computer etc. of (such as human-computer interaction is carried out by touch tablet), the mobile electricity
Sub- product can use any operating system, such as android operating system, iOS operating system.Wherein, the network equipment
The electronic equipment of numerical value calculating and information processing can be carried out automatically according to the instruction for being previously set or storing including a kind of,
Hardware includes but is not limited to microprocessor, specific integrated circuit (ASIC), programmable logic device (PLD), field programmable gate
Array (FPGA), digital signal processor (DSP), embedded device etc..The network equipment includes but is not limited to computer, net
The cloud that network host, single network server, multiple network server collection or multiple servers are constituted；Here, cloud is by based on cloud
The a large number of computers or network servers for calculating (Cloud Computing) is constituted, wherein cloud computing is the one of distributed computing
Kind, a virtual supercomputer consisting of a loosely coupled set of computers.The network includes but is not limited to interconnect
Net, wide area network, Metropolitan Area Network (MAN), local area network, VPN network, wireless self-organization network (Ad Hoc network) etc..Preferably, the equipment
Can also be run on the user equipment, the network equipment or user equipment and the network equipment, the network equipment, touch terminal or
The network equipment and touch terminal are integrated the program in constituted equipment by network.
Certainly, those skilled in the art will be understood that above equipment is only for example, other are existing or are likely to occur from now on
Equipment be such as applicable to the application, should also be included within the application protection scope, and be incorporated herein by reference.
In the description of the present application, the meaning of " plurality " is two or more, unless otherwise specifically defined.
Below based on a kind of equipment (or determine equipment for impact factor) for determining the impact factor of research object,
The application is described in detail.
According to the one aspect of the application, a kind of method for determining the impact factor of research object is provided.With reference to
Fig. 1, the method comprising the steps of S100, step S200 and step S300.Specifically, in the step s 100, impact factor determines
Equipment determines the cognitive appraisal information of the research object, pays close attention to evaluation information and enliven evaluation information.Wherein, in some realities
It applies in example, cognitive appraisal information is obtained based on the context analyzer to research object, for characterizing the overall popularity of research object,
It generally increases as the time increases, and variation is smaller in the short time；Pay close attention to the concerned journey of evaluation information image study object
Degree, alternatively referred to as temperature, the variation for cognitive appraisal information in the short time are possible to bigger sometimes；Actively comment
Valence information is then able to reflect the variation tendency (or being interpreted as the degree that temperature rises or falls) of concern evaluation information.In step
In S200, impact factor determines that equipment is based on the cognitive appraisal information, the concern evaluation information and the active evaluation
Information determines corresponding cognitive appraisal vector respectively, pays close attention to evaluation vector and enliven evaluation vector；Later in step S300
In, impact factor determines that equipment answers the cognitive appraisal vector, the concern evaluation vector and the evaluation vector that enlivens
The research pair is determined for the impact factor model about the research object, and based on the output of the impact factor model
As current impact factor.Here, impact factor model is a data model；In some embodiments, the impact factor model
Gained is developed based on deep neural network.Certainly, those skilled in the art will be understood that the impact factor model is not limited to base
Gained is developed in deep neural network, other data models that are existing or being likely to occur from now on such as can be suitably used for the application,
Comprising within the scope of protection of this application, and it is incorporated herein by reference.For example, the impact factor model is based on nerve net
One or more data models such as network, support vector machines (Support Vector Machine, SVM), decision tree, random forest
Exploitation gained.
For clarity, herein by taking the research object is a football player (hereinafter referred to as sportsman) as an example, to cognition
Evaluation information pays close attention to evaluation information, enlivens the various aspects such as evaluation information and be explained further.
One, cognitive appraisal information
Cognitive appraisal information is used to characterize the overall popularity of the sportsman, the overall popularity of in general one sportsman with
The time increase and increase, therefore reflect the degree of awareness of the sportsman in the minds of the public, varying less in the general short time.
In some embodiments, cognitive appraisal information Z can be calculated by following formula:
Z=w1*m1+w2*m2+…+wn*mn
Wherein, Z is popularity scoring, and m is the input of each data source, and w is the weight in each input data source；Each input
Data belong to the slower sportsman's background data of variation for simplicity can be by these attribution datas in " context analyzer module ".It needs
It should be noted that in some embodiments, the weight of each data source does not need explicitly to calculate herein；From these data
The data in source can be input to impact factor model, by impact factor model come the weight of each data source of adjust automatically so that whole
The weight of a context analyzer module, to obtain to the sportsman most accurately finally scoring (impact factor).More specifically, one
In a little embodiments, the cognitive appraisal information be based on research object at least one of attribute information determine therefore have certain
Objectivity, and change over time it is relatively slow, thus facilitate promoted final result accuracy.For example, context analyzer mould
May include the basic document of soccer star in block, for example including but be not limited to: it is nationality, effect club, annual pay, the effect time limit, leading
Position, adds up as number of national team's effect etc. at the main force/substitute information.
Two, it pays close attention to evaluation information and enlivens evaluation information
Concern evaluation information is used to capture the concerned situation (or active degree) in soccer star's short period, can be used for sending out
Existing hot spot sportsman, such as evaluation information will be paid close attention to and be used to measure the degree of sportsman's scoring change (for example, the scoring for enlivening sportsman changes
Time-varying amplitude may be bigger, and the scoring of sluggish sportsman is then relatively stable).Therefore, concern evaluation information also can be used
It is evaluated in sportsman, some embodiments are also using concern evaluation information as the input of impact factor model.With cognitive appraisal
Information is corresponding, which can be attributed to " temperature analysis module ", implements also and in " context analyzer module "
Input data it is similar.
In some embodiments, which is the associated media information determination based on research object, for example,
Associated media information include but is not limited to relevant to sportsman social network media information (including but not limited to picture, text,
Video content, and the data contents such as thumb up, forward), news media's information, search engine media information (including but not limited to
Ranking increment, entry increment) etc. in one or more.
Wherein, unlike " context analyzer module ", the data of part monitoring are incremental datas, such as in a timing
In, the quantity of sportsman's social networks newly-increased bean vermicelli quantity, or newly-increased related commentary, forwarding etc..It is paid close attention to by introducing
Evaluation information, we can grasp hot topic and hot spot sportsman, and these information are reflected into the shadow of corresponding sportsman in time
Ring the factor.Those skilled in the art will be understood that above-mentioned incremental data is only for example, other are existing or are likely to occur from now on
Incremental data such as can be suitably used for the application, be also contained in the protection scope of the application, and be incorporated herein by reference.One
In a little embodiments, above-mentioned steps S100 includes sub-step S110 (not shown), sub-step S120 (not shown) and sub-step S130
(not shown).In sub-step S110, impact factor determines that equipment determines the cognitive appraisal information of the research object；In sub-step
In rapid S120, impact factor determine equipment based on the research object at least one of associated media information determine it is corresponding at least
One increment media information and the research object enliven evaluation information, and wherein research object enlivens evaluation information
It is to be determined based at least one of described increment media information；In sub-step S130, it is described that impact factor determines that equipment is based on
At least one increment media information determines the concern evaluation information of the research object.
In some embodiments, in above-mentioned sub-step S130, impact factor determines that equipment is based at least one of described increment
Media information weight corresponding to media information and every increment media information determines the concern evaluation letter of the research object
Breath.For example, more being weighted during being monitored to specified country or global social network media or news media
The media report of prestige will obtain higher influence power weight, to make to the influence power of research object (such as sportsman in precedent)
The influence of Cheng Geng great；Can also the content of the discussions of the report to news media and social media carry out mood analysis, to obtain carriage
Whether be conducive to the influence power of a sportsman by guiding.These data can by the monitoring to social networks and search engine come
It obtains, such as club's data of sportsman and each social media is crawled using spiders.In some embodiments,
Above-mentioned sub-step S120 further includes sub-step S121 (not shown), sub-step S122 (not shown).In sub-step S121, influence
The factor determines that equipment determines corresponding increment media information based at least one associated media information of the research object, such as
Associated media information is obtained based on period regular hour, and corresponding increment media information is obtained according to more new information；In son
In step S122, impact factor determines that equipment executes sort operation at least one of described associated media information, and according to correspondence
Classification results determine the research object enliven evaluation information.
Wherein, above-mentioned sort operation is based on sorting algorithm execution in some embodiments, such as is held based on support vector machines
Row；Enlivening evaluation information can be used for analyzing the public to the mood of research object (such as sportsman in upper example), therefore corresponding defeated
" mood analysis module " can be included by entering data.The direction (such as liveness is positive or negative) of soccer star's active degree
It mood analysis module can be used to carry out control, the public obtained by the data content in analysis media and news to soccer star's
Emotion, such as by the user comment and forwarding data progress sentiment analysis to popular sportsman, learn that the public is to support the sportsman,
Still oppose the sportsman.Specifically, mood analysis is the text reported about a soccer star news media and social media
Analysis；By the analysis to text semanteme, the evaluation of news media and social media about a soccer star can be learnt, to count
Enter the entirety marking to the soccer star.The cardinal principle of support vector machines is: determining that optimal separation is different classes of in search space
Linear or nonlinear dividing strip, such as classification herein can be defined as " strong to support ", " support ", " neutrality ",
" opposition " and " being strongly opposed to " finally obtains masses to this ball based on commenting on or forwarding using supporting vector score of the game class each
The mood of star；The point of mistake classification this may be because of the ambiguity of the data or parameter of the vector machine model is very little or dimension
It spends too low.Support vector machines is a kind of machine learning algorithm for needing to supervise, so supporting vector machine model in order to obtain, needs
It is trained by certain data in advance.In these data (comment), specific classification mark is needed.Data volume
Size will directly determine the accuracy of supporting vector machine model.
The input of above-mentioned impact factor model, in addition to above-described cognitive appraisal information, concern evaluation information and work
It jumps except evaluation information etc., may also include historical data (such as the weighting of history score data about the research object
Average), hereinafter referred to as backtracking influences information.
Three, backtracking influences information
In some embodiments, in above-mentioned steps S300, impact factor determines equipment by the cognitive appraisal vector, institute
State concern evaluation vector, it is described enliven evaluation vector and about the research object backtracking influence vector, be applied to about
The impact factor model of the research object, and current based on the determining research object of the output of the impact factor model
Impact factor, to advantageously ensure that the stability and reliability of output data.Wherein, the backtracking influences vector based on backtracking
It influences to obtain after information carries out vectorization.By taking backtracking influences information for history scoring mean value as an example, it can be used simple but effective
Model is weighted and averaged the calculating of value based on following formula to history scoring:
λi=Σ W(i-j)λ’(i-j)(1≤j≤n)
Wherein λiIt is scoring of the i-th to a sportsman, λ '(i-j)It is (i-j) secondary history scoring record, above formula is from the
I scoring starts, and traces n times scoring forward；W(i-j)It is i-th scoring proportion (weight).It in some embodiments, is institute
There is history scoring to distribute identical weight.In further embodiments, forEachWeight carries out logarithm decline and calculates.Accordingly
Ground, above-mentioned steps S300 include sub-step S310, sub-step S320, sub-step S330 and sub-step S340.In sub-step S310
In, impact factor determines that equipment influences each backtracking at least one backtracking impact factor about the research object
The factor determines corresponding initial decline weight；In sub-step S320, impact factor determines equipment to the initial decline weight
Data normalization operation is executed, with backtracking impact factor weight corresponding to each backtracking impact factor of determination；In sub-step
In S330, impact factor determines equipment according to corresponding at least one described backtracking impact factor and each backtracking impact factor
Backtracking impact factor weight, determine that corresponding backtracking influences vector；In sub-step S340, impact factor determines equipment by institute
Stating cognitive appraisal vector, the concern evaluation vector, the evaluation vector and the backtracking of enlivening influences vector, is applied to close
Determine that the research object is current in the impact factor model of the research object, and based on the output of the impact factor model
Impact factor.For example, to obtain the corresponding backtracking impact factor weight W of history scorings, can count according to the following formula first
Calculate initial decline weight:
W’s=β ^ (i-s)
Wherein, β is fading parameter, is typically in the range of between 0 and 1.Then, by executing data standard to all scoring weights
Change operation, to obtain above-mentioned backtracking impact factor weight Ws.Calculating can be carried out based on following formula:
Ws=Ws’/ΣW’(i-j)(1≤j≤n)
Weighted average then based on the corresponding weight calculation history scoring of the scoring of above-mentioned history and the scoring of each history
Number, and determine that corresponding backtracking influences vector, then backtracking influence vector is inputted into impact factor mould together with other information together
Type, to obtain final output.Based on the above method, can neatly score from current scoring retrospect n, and to these score into
Row is flexibly and effectively weighted and averaged, and obtains a most suitable history scoring, in the points-scoring system, nearest score data will
Occupy bigger weight, thus helps to ensure that the stability and reliability of output data.
In some embodiments, above-mentioned impact factor model is generated based on deep neural network.For example, the influence because
Submodel uses input of the feature extracted as neural network, wherein we assume that each feature is by a bivector table
Show, the expression dimension of feature will be adjusted accordingly according to actual feature extraction result during specific implementation.At one
In specific embodiment, deep neural network is 7 layers of full Connection Neural Network, can be according to net in training process during practical realization
The degree of convergence and precision of network, optimize the structure of neural network, such as use convolutional neural networks, coding and decoding nerve
Network etc., so that output result is more accurate.In neural network, pass through matrix operation and activation equation between neuron
The power of signal transmitting is controlled, can learn to most perfect feature to use and assembled scheme.In this application, above-mentioned
The each feature of impact factor model reasonable distribution neural network based (such as mould is analyzed from context analyzer module, temperature
The data of block, mood analysis module, history scoring etc.) to the influence degree of output result (such as this scoring).This field
Technical staff will be understood that above-mentioned impact factor model is only for example, other existing or influences for being likely to occur from now on because
Submodel such as can be suitably used for the application, be also contained in the protection scope of the application, and be incorporated herein by reference.
In order to export the accuracy of result, Yi Xieshi with optimization according to the public above-mentioned impact factor model of feedback adjustment
It applies example and feedback analysis is carried out to field feedback.Correspondingly, the above method further includes step S400 (not shown), step S500
(not shown) and step S600 (not shown).In step S400, impact factor determines that equipment obtains at least one target user
Field feedback about the current impact factor；In step S500, impact factor determines that equipment is determined about institute
State the feedback intensity information and feedback emotion information of field feedback；In step S600, impact factor determines equipment base
In the feedback intensity information, the feedback emotion information and the current impact factor, the impact factor mould is determined
The loss of type, and the impact factor model is updated based on the loss.It is opened with above-mentioned impact factor model based on neural network
For hair, in some embodiments, specifically, user data is monitored first, to feedback quantity (feedback intensity) and
It feeds back emotion (positive feedback or negative-feedback) and carries out data preparation and analysis.For feedback intensity, feedback data passes through standardization
Processing is established and using linear or multidimensional regression model, it is established that the relationship of data volume and feedback intensity, thus by data
Amount is converted into feedback intensity；For feeding back emotion, we will be used with mood analysis method described above, model the feelings of user
Thread data.User is finally obtained in this way to the field feedback of scoring, a kind of form of expression is that user thinks that accuracy is
A%, wherein b% user is positive, that is, thinks that scoring should increase.This data forms the output with neural network
Loss then updates the numerical value of parameters in neural network, such as the reasonable mind of distribution using the backpropagation of neural network
Through e-learning rate (Learning Rate), so that neural network is absorbed the feedback and make corresponding adjustment, is updated with exporting
Score data afterwards, which considers feedback information of the model with user itself, so that neural network court
More accurately direction develop.It should be noted that above-mentioned feedback system needs to be arranged when using Neural Network Data model
Reasonable neural network learning rate, too high learning rate will lead to fluctuating widely for neural network scoring, to lose
Versatility and accuracy；Too small learning rate can make the scoring feedback of user not impact to the output of neural network.
Wherein, it should be noted that be only for example above based on related description of the football player to the application, without right
The application field of the application carries out any restriction.In fact, based on the system to score constructed by above-mentioned technology sportsman
It can also be used for other entities, such as the temperature of a video display star or block chain project evaluated.For example, based on above-mentioned
System design, reusable following parameter:
Feature extraction；
Neural network framework；
Score feedback system framework.
In characteristic extraction procedure, the crawler for tracking sportsman need to only be replaced with to tracking video display star or block chain project
Crawler, adjust corresponding data and crawl source, the data of corresponding star or block chain project can be obtained, then by similar
Algorithm, context analyzer information, temperature analysis information, sentiment analysis information, the history for obtaining corresponding star or block chain project are flat
The features such as weighting.Then, the feature extracted can be input in neural network, is debugged by different neural networks,
Scoring is optimized.Finally, scoring will enter into scoring feedback system, the correction scored by user group/expert.But
It should be noted that in order to which whole system can preferably be commented for another entity (such as star or block chain project)
Point, it may be necessary to adjust/redesign following modules:
Data extraction source；
Data processing dictionary；
Neural network debugging；
Feedback system debugging.
In short, the above constructed evaluation that can extend to for the points-scoring system of sportsman's temperature for other entities
Cheng Zhong, therefore the scheme of the impact factor for determining research object provided herein has stronger scalability.
According to further aspect of the application, a kind of equipment for determining the impact factor of research object is provided.Ginseng
Fig. 2 is examined, which includes the first module 100, the second module 200 and third module 300.Specifically, the first module 100 determines
The cognitive appraisal information of the research object pays close attention to evaluation information and enlivens evaluation information.Wherein, in some embodiments,
Cognitive appraisal information is obtained based on the context analyzer to research object, general for characterizing the overall popularity of research object
Increase as the time increases, variation is smaller in the short time；The concerned degree of evaluation information image study object is paid close attention to, sometimes
Alternatively referred to as temperature, the variation for cognitive appraisal information in the short time is possible to can be bigger；Enliven evaluation information
Then it is able to reflect the variation tendency (or being interpreted as the degree that temperature rises or falls) of concern evaluation information.Second module, 200 base
In the cognitive appraisal information, the concern evaluation information and it is described enliven evaluation information, determine respectively it is corresponding cognition comment
Valence vector pays close attention to evaluation vector and enlivens evaluation vector；Third module 300 is by the cognitive appraisal vector, the pass later
Note evaluation vector and the evaluation vector that enlivens are based on described applied to the impact factor model about the research object
The output of impact factor model determines the current impact factor of the research object.Here, impact factor model is a data mould
Type；In some embodiments, which is based on deep neural network exploitation gained.Certainly, those skilled in the art
It will be understood that the impact factor model is not limited to based on deep neural network exploitation gained, other are existing or from now on may
The data model of appearance such as can be suitably used for the application, be also contained in the protection scope of the application, and be contained in by reference
This.For example, the impact factor model is based on neural network, support vector machine (Support Vector Machine, SVM), determines
One or more data model exploitation gained such as plan tree, random forest.
For clarity, herein by taking the research object is a football player (hereinafter referred to as sportsman) as an example, to cognition
Evaluation information pays close attention to evaluation information, enlivens the various aspects such as evaluation information and be explained further.
One, cognitive appraisal information
Cognitive appraisal information is used to characterize the overall popularity of the sportsman, the overall popularity of in general one sportsman with
The time increase and increase, therefore reflect the degree of awareness of the sportsman in the minds of the public, varying less in the general short time.
In some embodiments, cognitive appraisal information Z can be calculated by following formula:
Z=w1*m1+w2*m2+…+wn*mn
Wherein, Z is popularity scoring, and m is the input of each data source, and w is the weight in each input data source；Each input
Data belong to the slower sportsman's background data of variation for simplicity can be by these attribution datas in " context analyzer module ".It needs
It should be noted that in some embodiments, the weight of each data source does not need explicitly to calculate herein；From these data
The data in source can be input to impact factor model, by impact factor model come the weight of each data source of adjust automatically so that whole
The weight of a context analyzer module, to obtain to the sportsman most accurately finally scoring (impact factor).More specifically, one
In a little embodiments, the cognitive appraisal information be based on research object at least one of attribute information determine therefore have certain
Objectivity, and change over time it is relatively slow, thus facilitate promoted final result accuracy.For example, context analyzer mould
May include the basic document of soccer star in block, for example including but be not limited to: it is nationality, effect club, annual pay, the effect time limit, leading
Position, adds up as number of national team's effect etc. at the main force/substitute information.
Two, it pays close attention to evaluation information and enlivens evaluation information
Concern evaluation information is used to capture the concerned situation (or active degree) in soccer star's short period, can be used for sending out
Existing hot spot sportsman, such as evaluation information will be paid close attention to and be used to measure the degree of sportsman's scoring change (for example, the scoring for enlivening sportsman changes
Time-varying amplitude may be bigger, and the scoring of sluggish sportsman is then relatively stable).Therefore, concern evaluation information also can be used
It is evaluated in sportsman, some embodiments are also using concern evaluation information as the input of impact factor model.With cognitive appraisal
Information is corresponding, which can be attributed to " temperature analysis module ", implements also and in " context analyzer module "
Input data it is similar.
In some embodiments, which is the associated media information determination based on research object, for example,
Associated media information include but is not limited to relevant to sportsman social network media information (including but not limited to picture, text,
Video content, and the data contents such as thumb up, forward), news media's information, search engine media information (including but not limited to
Ranking increment, entry increment) etc. in one or more.
Wherein, unlike " context analyzer module ", the data of part monitoring are incremental datas, such as in a timing
In, the quantity of sportsman's social networks newly-increased bean vermicelli quantity, or newly-increased related commentary, forwarding etc..It is paid close attention to by introducing
Evaluation information, we can grasp hot topic and hot spot sportsman, and these information are reflected into the shadow of corresponding sportsman in time
Ring the factor.Those skilled in the art will be understood that above-mentioned incremental data is only for example, other are existing or are likely to occur from now on
Incremental data such as can be suitably used for the application, be also contained in the protection scope of the application, and be incorporated herein by reference.One
In a little embodiments, above-mentioned first module 100 includes 110 (not shown) of first unit, 120 (not shown) of second unit and third list
First 130 (not shown).First unit 110 determines the cognitive appraisal information of the research object；Second unit 120 is based on described grind
Study carefully object at least one of associated media information determine at least one of corresponding increment media information and the research object
Evaluation information is enlivened, wherein the evaluation information that enlivens of research object is also based at least one of described increment media information determination
's；Third unit 130 determines the concern evaluation information of the research object based at least one of described increment media information.
In some embodiments, above-mentioned third unit 130 is based at least one of described increment media information and every increasing
Media information weight corresponding to media information is measured, determines the concern evaluation information of the research object.For example, to designated state
During the social network media or news media in family or the whole world are monitored, more authoritative media report will be obtained
Higher influence power weight, to cause bigger influence to the influence power of research object (such as sportsman in precedent)；May be used also
The content of the discussions of report and social media to news media carries out mood analysis, to obtain whether spin is conducive to one
The influence power of a sportsman.These data can be obtained by the monitoring to social networks and search engine, such as use webpage
Crawler crawls club's data of sportsman and each social media.In some embodiments, above-mentioned second unit 120 is gone back
Including 121 (not shown) of first assembly, 122 (not shown) of the second component.First assembly 121 based on the research object at least
One associated media information determines corresponding increment media information, such as obtains related media letter based on period regular hour
Breath, and corresponding increment media information is obtained according to more new information；Second component 122 is at least one of described associated media information
Sort operation is executed, and enlivens evaluation information according to what corresponding classification results determined the research object.
Wherein, above-mentioned sort operation is based on sorting algorithm execution in some embodiments, such as is held based on support vector machines
Row；Enlivening evaluation information can be used for analyzing the public to the mood of research object (such as sportsman in upper example), therefore corresponding defeated
" mood analysis module " can be included by entering data.The direction (such as liveness is positive or negative) of soccer star's active degree
It mood analysis module can be used to carry out control, the public obtained by the data content in analysis media and news to soccer star's
Emotion, such as by the user comment and forwarding data progress sentiment analysis to popular sportsman, learn that the public is to support the sportsman,
Still oppose the sportsman.Specifically, mood analysis is the text reported about a soccer star news media and social media
Analysis；By the analysis to text semanteme, the evaluation of news media and social media about a soccer star can be learnt, to count
Enter the entirety marking to the soccer star.The cardinal principle of support vector machines is: determining that optimal separation is different classes of in search space
Linear or nonlinear dividing strip, such as classification herein can be defined as " strong to support ", " support ", " neutrality ",
" opposition " and " being strongly opposed to " finally obtains masses to this ball based on commenting on or forwarding using supporting vector score of the game class each
The mood of star；The point of mistake classification this may be because of the ambiguity of the data or parameter of the vector machine model is very little or dimension
It spends too low.Support vector machines is a kind of machine learning algorithm for needing to supervise, so supporting vector machine model in order to obtain, needs
It is trained by certain data in advance.In these data (comment), specific classification mark is needed.Data volume
Size will directly determine the accuracy of supporting vector machine model.
The input of above-mentioned impact factor model, in addition to above-described cognitive appraisal information, concern evaluation information and work
It jumps except evaluation information etc., may also include historical data (such as the weighting of history score data about the research object
Average), hereinafter referred to as backtracking influences information.
Three, backtracking influences information
In some embodiments, above-mentioned third module 300 is by the cognitive appraisal vector, the concern evaluation vector, institute
It states and enlivens evaluation vector and the backtracking influence vector about the research object, applied to the influence about the research object
Factor model, and the current impact factor of the research object is determined based on the output of the impact factor model, thus favorably
In the stability and reliability that guarantee output data.Wherein, the backtracking, which influences vector, influences information progress vector based on backtracking
It is obtained after change.By backtracking influence information be history score mean value for, simple but effective model can be used, based on following formula come pair
History, which scores, is weighted and averaged the calculating of value:
λi=Σ W(i-j)λ’(i-j)(1≤j≤n)
Wherein λiIt is scoring of the i-th to a sportsman, λ '(i-j)It is (i-j) secondary history scoring record, above formula is from the
I scoring starts, and traces n times scoring forward；W(i-j)It is i-th scoring proportion (weight).It in some embodiments, is institute
There is history scoring to distribute identical weight.In further embodiments, logarithm decline is carried out for each weight to calculate.Accordingly
Ground, above-mentioned third module 300 include the 4th unit 310, the 5th unit 320, the 6th unit 330 and the 7th unit 340.4th is single
Member 310 is determined and is corresponded to for each backtracking impact factor at least one backtracking impact factor about the research object
Initial decline weight；5th unit 320 executes data normalization operation to the initial decline weight, with each backtracking of determination
Backtracking impact factor weight corresponding to impact factor；6th unit 330 according to it is described at least one backtracking impact factor and
Backtracking impact factor weight corresponding to each backtracking impact factor determines that corresponding backtracking influences vector；7th unit 340 will
The cognitive appraisal vector, the concern evaluation vector, the evaluation vector and the backtracking of enlivening influence vector, are applied to
Determine that the research object is worked as about the impact factor model of the research object, and based on the output of the impact factor model
Preceding impact factor.For example, to obtain the corresponding backtracking impact factor weight W of history scorings, can first according to the following formula
Calculate initial decline weight:
W’s=β ^ (i-s)
Wherein, β is fading parameter, is typically in the range of between 0 and 1.Then, by executing data standard to all scoring weights
Change operation, to obtain above-mentioned backtracking impact factor weight Ws.Calculating can be carried out based on following formula:
Ws=Ws’/ΣW’(i-j)(1≤j≤n)
Weighted average then based on the corresponding weight calculation history scoring of the scoring of above-mentioned history and the scoring of each history
Number, and determine that corresponding backtracking influences vector, then backtracking influence vector is inputted into impact factor mould together with other information together
Type, to obtain final output.Based on the above method, can neatly score from current scoring retrospect n, and to these score into
Row is flexibly and effectively weighted and averaged, and obtains a most suitable history scoring, in the points-scoring system, nearest score data will
Occupy bigger weight, thus helps to ensure that the stability and reliability of output data.
In some embodiments, above-mentioned impact factor model is generated based on deep neural network.For example, the influence because
Submodel uses input of the feature extracted as neural network, wherein we assume that each feature is by a bivector table
Show, the expression dimension of feature will be adjusted accordingly according to actual feature extraction result during specific implementation.At one
In specific embodiment, deep neural network is 7 layers of full Connection Neural Network, can be according to net in training process during practical realization
The degree of convergence and precision of network, optimize the structure of neural network, such as use convolutional neural networks, coding and decoding nerve
Network etc., so that output result is more accurate.In neural network, pass through matrix operation and activation equation between neuron
The power of signal transmitting is controlled, can learn to most perfect feature to use and assembled scheme.In this application, above-mentioned
The each feature of impact factor model reasonable distribution neural network based (such as mould is analyzed from context analyzer module, temperature
The data of block, mood analysis module, history scoring etc.) to the influence degree of output result (such as this scoring).This field
Technical staff will be understood that above-mentioned impact factor model is only for example, other existing or influences for being likely to occur from now on because
Submodel such as can be suitably used for the application, be also contained in the protection scope of the application, and be incorporated herein by reference.
In order to export the accuracy of result, Yi Xieshi with optimization according to the public above-mentioned impact factor model of feedback adjustment
It applies example and feedback analysis is carried out to field feedback.Correspondingly, above equipment further includes 400 (not shown) of the 4th module, the 5th
500 (not shown) of module and 600 (not shown) of the 6th module.4th module 400 obtains at least one target user about described
The field feedback of current impact factor；5th module 500 determines that the feedback intensity about the field feedback is believed
Breath and feedback emotion information；6th module 600 is based on the feedback intensity information, the feedback emotion information and described works as
Preceding impact factor determines the loss of the impact factor model, and updates the impact factor model based on the loss.With
Above-mentioned impact factor model be based on development of neural networks for, in some embodiments, specifically, first to user data into
Row monitoring carries out data preparation and analysis to feedback quantity (feedback intensity) and feedback emotion (positive feedback or negative-feedback).It is right
In feedback intensity, feedback data is established and using linear or multidimensional regression model, it is established that data by standardization
The relationship of amount and feedback intensity, to convert feedback intensity for data volume；For feeding back emotion, we will be used with Jie above
The mood analysis method to continue, models the mood data of user.User is finally obtained in this way to the field feedback of scoring, one kind
The form of expression is that user thinks that accuracy is a%, and wherein b% user is positive, that is, thinks that scoring should increase.This data
Output with neural network is formed into loss, then updates parameters in neural network using the backpropagation of neural network
Numerical value, such as reasonable neural network learning rate (Learning Rate) is distributed, so that neural network is absorbed the feedback and makes
Corresponding adjustment, to export updated score data, which considers the anti-of model itself and user
Feedforward information, so that neural network develops towards more accurate direction.It should be noted that using Neural Network Data model
When, above-mentioned feedback system needs to be arranged reasonable neural network learning rate, and too high learning rate will lead to neural network and comment
That divides fluctuates widely, to lose versatility and accuracy；It is not right that too small learning rate can make the scoring of user feed back
The output of neural network impacts.
Wherein, it should be noted that be only for example above based on related description of the football player to the application, without right
The application field of the application carries out any restriction.In fact, based on the system to score constructed by above-mentioned technology sportsman
It can also be used for other entities, such as the temperature of a video display star or block chain project evaluated.For example, based on above-mentioned
System design, reusable following parameter:
Feature extraction；
Neural network framework；
Score feedback system framework.
In characteristic extraction procedure, the crawler for tracking sportsman need to only be replaced with to tracking video display star or block chain project
Crawler, adjust corresponding data and crawl source, the data of corresponding star or block chain project can be obtained, then by similar
Algorithm, context analyzer information, temperature analysis information, sentiment analysis information, the history for obtaining corresponding star or block chain project are flat
The features such as weighting.Then, the feature extracted can be input in neural network, is debugged by different neural networks,
Scoring is optimized.Finally, scoring will enter into scoring feedback system, the correction scored by user group/expert.But
It should be noted that in order to which whole system can preferably be commented for another entity (such as star or block chain project)
Point, it may be necessary to adjust/redesign following modules:
Data extraction source；
Data processing dictionary；
Neural network debugging；
Feedback system debugging.
In short, the above constructed evaluation that can extend to for the points-scoring system of sportsman's temperature for other entities
Cheng Zhong, therefore the scheme of the impact factor for determining research object provided herein has stronger scalability.
Present invention also provides a kind of computer readable storage medium, the computer-readable recording medium storage has calculating
Machine code, when the computer code is performed, such as preceding described in any item methods are performed.
Present invention also provides a kind of computer program products, when the computer program product is executed by computer equipment
When, such as preceding described in any item methods are performed.
Present invention also provides a kind of computer equipment, the computer equipment includes:
One or more processors；
Memory, for storing one or more computer programs；
When one or more of computer programs are executed by one or more of processors so that it is one or
Multiple processors realize such as preceding described in any item methods.
Fig. 3 shows the exemplary system that can be used for implementing each embodiment described herein.
As shown in figure 3, in some embodiments, system 700 can be influenced as any one in each embodiment
The factor determines equipment.In some embodiments, system 700 may include one or more computer-readable mediums with instruction
It (for example, system storage or NVM/ store equipment 720) and couples and is matched with the one or more computer-readable medium
Be set to execute instruction with realize module thereby executing movement described herein one or more processors (for example, (one
Or multiple) processor 705).
For one embodiment, system control module 710 may include any suitable interface controller, with to (one or
It is multiple) at least one of processor 705 and/or any suitable equipment or component that communicate with system control module 710 mentions
For any suitable interface.
System control module 710 may include Memory Controller module 730, to provide interface to system storage 715.It deposits
Memory controller module 730 can be hardware module, software module and/or firmware module.
System storage 715 can be used for for example, load of system 700 and storing data and/or instruction.For a reality
Example is applied, system storage 715 may include any suitable volatile memory, for example, DRAM appropriate.In some embodiments
In, system storage 715 may include four Synchronous Dynamic Random Access Memory of Double Data Rate type (DDR4SDRAM).
For one embodiment, system control module 710 may include one or more input/output (I/O) controller, with
Equipment 720 is stored to NVM/ and (one or more) communication interface 725 provides interface.
For example, NVM/ storage equipment 720 can be used for storing data and/or instruction.NVM/ storage equipment 720 may include appointing
It anticipates nonvolatile memory appropriate (for example, flash memory) and/or to may include that any suitable (one or more) is non-volatile deposit
Equipment is stored up (for example, one or more hard disk drives (HDD), one or more CD (CD) drivers and/or one or more
Digital versatile disc (DVD) driver).
NVM/ storage equipment 720 may include a part for the equipment being physically mounted on as system 700
Storage resource or its can by the equipment access without a part as the equipment.For example, NVM/ storage equipment 720 can
It is accessed by network via (one or more) communication interface 725.
(one or more) communication interface 725 can be provided for system 700 interface with by one or more networks and/or with
Other any equipment communications appropriate.System 700 can be according to any mark in one or more wireless network standards and/or agreement
Quasi- and/or agreement is carried out wireless communication with the one or more components of wireless network.
For one embodiment, at least one of (one or more) processor 705 can be with system control module 710
The logic of one or more controllers (for example, Memory Controller module 730) is packaged together.For one embodiment, (one
It is a or multiple) at least one of processor 705 can encapsulate with the logic of one or more controllers of system control module 710
Together to form system in package (SiP).For one embodiment, at least one of (one or more) processor 705
It can be integrated on same mold with the logic of one or more controllers of system control module 710.For one embodiment,
At least one of (one or more) processor 705 can be with the logic of one or more controllers of system control module 710
It is integrated on same mold to form system on chip (SoC).
In various embodiments, system 700 can be, but not limited to be: server, work station, desk-top calculating equipment or movement
It calculates equipment (for example, lap-top computing devices, handheld computing device, tablet computer, net book etc.).In various embodiments,
System 700 can have more or fewer components and/or different frameworks.For example, in some embodiments, system 700 includes
One or more video cameras, keyboard, liquid crystal display (LCD) screen (including touch screen displays), nonvolatile memory port,
Mutiple antennas, graphic chips, specific integrated circuit (ASIC) and loudspeaker.
It should be noted that the application can be carried out in the assembly of software and/or software and hardware, for example, can adopt
With specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment
In, the software program of the application can be executed to implement the above steps or functions by processor.Similarly, the application
Software program (including relevant data structure) can be stored in computer readable recording medium, for example, RAM memory,
Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the application, example
Such as, as the circuit cooperated with processor thereby executing each step or function.
In addition, a part of the application can be applied to computer program product, such as computer program instructions, when its quilt
When computer executes, by the operation of the computer, it can call or provide according to the present processes and/or technical solution.
Those skilled in the art will be understood that the existence form of computer program instructions in computer-readable medium includes but is not limited to
Source file, executable file, installation package file etc., correspondingly, the mode that computer program instructions are computer-executed include but
Be not limited to: the computer directly execute the instruction or the computer compile the instruction after execute program after corresponding compiling again,
Perhaps the computer reads and executes the instruction or after the computer reads and install and execute corresponding installation again after the instruction
Program.Here, computer-readable medium can be for computer access any available computer readable storage medium or
Communication media.
Communication media includes whereby including, for example, computer readable instructions, data structure, program module or other data
Signal of communication is transmitted to the medium of another system from a system.Communication media may include having the transmission medium led (such as electric
Cable and line (for example, optical fiber, coaxial etc.)) and can propagate wireless (not having the transmission the led) medium of energy wave, such as sound, electricity
Magnetic, RF, microwave and infrared.Computer readable instructions, data structure, program module or other data can be embodied as example wireless
Medium (such as carrier wave or be such as embodied as spread spectrum technique a part similar mechanism) in modulated message signal.
Term " modulated message signal " refers to that one or more feature is modified or is set in a manner of encoded information in the signal
Fixed signal.Modulation can be simulation, digital or Hybrid Modulation Technology.
As an example, not a limit, computer readable storage medium may include such as computer-readable finger for storage
Enable, the volatile and non-volatile that any method or technique of the information of data structure, program module or other data is realized, can
Mobile and immovable medium.For example, computer readable storage medium includes, but are not limited to volatile memory, such as with
Machine memory (RAM, DRAM, SRAM)；And nonvolatile memory, such as flash memory, various read-only memory (ROM, PROM,
EPROM, EEPROM), magnetic and ferromagnetic/ferroelectric memory (MRAM, FeRAM)；And magnetic and optical storage apparatus (hard disk,
Tape, CD, DVD)；Or other currently known media or Future Development can store the computer used for computer system
Readable information/data.
Here, including a device according to one embodiment of the application, which includes for storing computer program
The memory of instruction and processor for executing program instructions, wherein when the computer program instructions are executed by the processor
When, trigger method and/or technology scheme of the device operation based on aforementioned multiple embodiments according to the application.
It is obvious to a person skilled in the art that the application is not limited to the details of above-mentioned exemplary embodiment, Er Qie
In the case where without departing substantially from spirit herein or essential characteristic, the application can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and scope of the present application is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included in the application.Any reference signs in the claims should not be construed as limiting the involved claims.This
Outside, it is clear that one word of " comprising " does not exclude other units or steps, and odd number is not excluded for plural number.That states in device claim is multiple
Unit or device can also be implemented through software or hardware by a unit or device.The first, the second equal words are used to table
Show title, and does not indicate any particular order.
Claims (24)
1. a kind of method for determining the impact factor of research object, wherein method includes the following steps:
A determines the cognitive appraisal information of the research object, pays close attention to evaluation information and enliven evaluation information；
B be based on the cognitive appraisal information, the concern evaluation information and it is described enliven evaluation information, determine respectively corresponding
Cognitive appraisal vector pays close attention to evaluation vector and enlivens evaluation vector；
The cognitive appraisal vector, the concern evaluation vector and the evaluation vector that enlivens are applied to grind about described by c
Study carefully the impact factor model of object, and based on the output of the impact factor model determine the current influence of the research object because
Son.
2. according to the method described in claim 1, wherein, the cognitive appraisal information is at least one based on the research object
Item attribute information determines.
3. method according to claim 1 or 2, wherein the concern evaluation information and the evaluation information that enlivens are
It is determined based on the associated media information of the research object.
4. according to the method described in claim 3, wherein, the associated media information includes following at least any one:
Social network media information；
News media's information；
Search engine media information.
5. according to the method described in claim 3, wherein, the step a includes:
A1 determines the cognitive appraisal information of the research object；
A2 determines at least one of corresponding increment media information based at least one associated media information of the research object, with
And the research object enlivens evaluation information；
A3 determines the concern evaluation information of the research object based at least one of described increment media information.
6. according to the method described in claim 5, wherein, the step a3 includes:
Based on media information weight corresponding at least one of described increment media information and every increment media information, determine
The concern evaluation information of the research object.
7. according to the method described in claim 5, wherein, the step a2 includes:
Corresponding increment media information is determined based at least one associated media information of the research object；
Sort operation is executed at least one of described associated media information, and determines the research pair according to corresponding classification results
Elephant enlivens evaluation information.
8. according to the method described in claim 1, wherein, the step c includes:
By the cognitive appraisal vector, the concern evaluation vector, described enliven evaluation vector and about the research object
Backtracking influence vector, applied to the impact factor model about the research object, and based on the impact factor model
Output determines the current impact factor of the research object.
9. according to the method described in claim 8, wherein, the step c includes:
For each backtracking impact factor at least one backtracking impact factor about the research object, determine corresponding
Initial decline weight；
Data normalization operation is executed to the initial decline weight, with backtracking shadow corresponding to each backtracking impact factor of determination
Ring Factor Weight；
According to it is described at least one backtracking impact factor and each backtracking impact factor corresponding to backtracking impact factor weight,
Determine that corresponding backtracking influences vector；
The cognitive appraisal vector, the concern evaluation vector, the evaluation vector and the backtracking of enlivening are influenced into vector,
The research is determined applied to the impact factor model about the research object, and based on the output of the impact factor model
The current impact factor of object.
10. according to the method described in claim 1, wherein, the method also includes:
Obtain field feedback of at least one target user about the current impact factor；
Determine the feedback intensity information and feedback emotion information about the field feedback；
Based on the feedback intensity information, the feedback emotion information and the current impact factor, the influence is determined
The loss of factor model, and the impact factor model is updated based on the loss.
11. according to the method described in claim 1, wherein, the impact factor model is generated based on deep neural network.
12. a kind of equipment for determining the impact factor of research object, wherein the equipment includes:
First module, for determining the cognitive appraisal information of the research object, paying close attention to evaluation information and enlivening evaluation information；
Second module, for based on the cognitive appraisal information, the concern evaluation information and it is described enliven evaluation information, divide
Corresponding cognitive appraisal vector is not determined, is paid close attention to evaluation vector and is enlivened evaluation vector；
Third module, for by the cognitive appraisal vector, the concern evaluation vector and described enlivening evaluation vector application
The research object is determined in the impact factor model about the research object, and based on the output of the impact factor model
Current impact factor.
13. equipment according to claim 12, wherein it is described wherein, the cognitive appraisal information is based on the research
What at least one attribute information of object determined.
14. equipment according to claim 12 or 13, wherein the concern evaluation information and described enliven evaluation information
It is the associated media information determination based on the research object.
15. equipment according to claim 14, wherein the associated media information includes following at least any one:
Social network media information；
News media's information；
Search engine media information.
16. equipment according to claim 14, wherein first module includes:
First unit, for determining the cognitive appraisal information of the research object；
Second unit determines at least one of corresponding increment at least one associated media information based on the research object
Media information and the research object enliven evaluation information；
Third unit, for determining the concern evaluation information of the research object based at least one of described increment media information.
17. equipment according to claim 16, wherein the third unit is used for:
Based on media information weight corresponding at least one of described increment media information and every increment media information, determine
The concern evaluation information of the research object.
18. equipment according to claim 16, wherein the second unit includes:
First assembly determines that corresponding increment media are believed at least one associated media information based on the research object
Breath；
Second component, for executing sort operation at least one of described associated media information, and according to corresponding classification results
Determine the research object enlivens evaluation information.
19. equipment according to claim 12, wherein the third module is used for:
By the cognitive appraisal vector, the concern evaluation vector, described enliven evaluation vector and about the research object
Backtracking influence vector, applied to the impact factor model about the research object, and based on the impact factor model
Output determines the current impact factor of the research object.
20. equipment according to claim 19, wherein the third module includes:
Unit the 4th, for for about the research object at least one backtracking impact factor in each backtracking influence because
Son determines corresponding initial decline weight；
Unit the 5th, for executing data normalization operation to the initial decline weight, with each backtracking impact factor of determination
Corresponding backtracking impact factor weight；
Unit the 6th, for the backtracking according to corresponding at least one described backtracking impact factor and each backtracking impact factor
Impact factor weight determines that corresponding backtracking influences vector；
Unit the 7th, for by the cognitive appraisal vector, the concern evaluation vector, the evaluation vector and described of enlivening
Backtracking influences vector, applied to the impact factor model about the research object, and based on the defeated of the impact factor model
The current impact factor of the research object is determined out.
21. equipment according to claim 12, wherein the equipment further include:
4th module, for obtaining field feedback of at least one target user about the current impact factor；
5th module, for determining feedback intensity information and feedback emotion information about the field feedback；
6th module, for based on the feedback intensity information, the feedback emotion information and the current impact factor,
It determines the loss of the impact factor model, and the impact factor model is updated based on the loss.
22. equipment according to claim 12, wherein the impact factor model is generated based on deep neural network.
23. a kind of equipment for determining the impact factor of research object, wherein the equipment includes:
Processor；And
It is arranged to the memory of storage computer executable instructions, the executable instruction makes the processor when executed
It executes according to claim 1 to the operation of any one of 11 the methods.
24. a kind of computer-readable medium including instruction, described instruction wants system execution according to right
Ask the operation of any one of 1 to 11 the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811265351.XA CN109447462A (en) | 2018-10-29 | 2018-10-29 | It is a kind of for determining the method and apparatus of the impact factor of research object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811265351.XA CN109447462A (en) | 2018-10-29 | 2018-10-29 | It is a kind of for determining the method and apparatus of the impact factor of research object |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109447462A true CN109447462A (en) | 2019-03-08 |
Family
ID=65549035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811265351.XA Pending CN109447462A (en) | 2018-10-29 | 2018-10-29 | It is a kind of for determining the method and apparatus of the impact factor of research object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447462A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263234A (en) * | 2019-05-06 | 2019-09-20 | 平安科技（深圳）有限公司 | Enterpriser's influence power appraisal procedure, device, computer equipment and storage medium |
| CN110503332A (en) * | 2019-08-21 | 2019-11-26 | 谷元(上海)文化科技有限责任公司 | A kind of information sifting processing method |
| TWI752546B (en) * | 2020-07-09 | 2022-01-11 | 多利曼股份有限公司 | Evaluation system and evaluation method |
-
2018
- 2018-10-29 CN CN201811265351.XA patent/CN109447462A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263234A (en) * | 2019-05-06 | 2019-09-20 | 平安科技（深圳）有限公司 | Enterpriser's influence power appraisal procedure, device, computer equipment and storage medium |
| CN110503332A (en) * | 2019-08-21 | 2019-11-26 | 谷元(上海)文化科技有限责任公司 | A kind of information sifting processing method |
| CN110503332B (en) * | 2019-08-21 | 2022-12-30 | 谷元(上海)文化科技有限责任公司 | Information screening processing method |
| TWI752546B (en) * | 2020-07-09 | 2022-01-11 | 多利曼股份有限公司 | Evaluation system and evaluation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rein et al. | Big data and tactical analysis in elite soccer: future challenges and opportunities for sports science | |
| Shen et al. | An ensemble method based on selection using bat algorithm for intrusion detection | |
| Jiang et al. | A new binary hybrid particle swarm optimization with wavelet mutation | |
| US11551284B2 (en) | Session-based recommendation method and device | |
| CN109214436A (en) | A kind of prediction model training method and device for target scene | |
| CN109447462A (en) | It is a kind of for determining the method and apparatus of the impact factor of research object | |
| Wang et al. | A restart univariate estimation of distribution algorithm: sampling under mixed Gaussian and Lévy probability distribution | |
| Li et al. | Exploring n-gram features in clickstream data for MOOC learning achievement prediction | |
| US20140188928A1 (en) | Relational database management | |
| Whitehead et al. | The use of technical-tactical and physical performance indicators to classify between levels of match-play in elite rugby league | |
| Xiao | Artificial intelligence programming with Python: from zero to hero | |
| Zhang | [Retracted] Automatic Detection Method of Technical and Tactical Indicators for Table Tennis Based on Trajectory Prediction Using Compensation Fuzzy Neural Network | |
| Wu et al. | LR-SVM+: Learning using privileged information with noisy labels | |
| US20220358589A1 (en) | Electronic trading platform | |
| Milano et al. | Automated curriculum learning for embodied agents a neuroevolutionary approach | |
| CN109299459A (en) | A kind of the term vector training method and device of single semantic supervision | |
| Hu et al. | Application of Teaching Quality Assessment Based on Parallel Genetic Support Vector Algorithm in the Cloud Computing Teaching System. | |
| CN110135592A (en) | Classifying quality determines method, apparatus, intelligent terminal and storage medium | |
| Zhang et al. | Artificial Intelligence‐Based Joint Movement Estimation Method for Football Players in Sports Training | |
| Herrmannova et al. | Evaluating weekly predictions of at-risk students at the open university: results and issues | |
| Liu et al. | Data analysis of the development status of Basketball National Fitness based on FOG Computing | |
| Yao et al. | Youth sports special skills’ training and evaluation system based on machine learning | |
| Al Maksur et al. | MyBotS Prototype on Social Media Discord with NLP | |
| KR102477565B1 (en) | Method and system for predicting reporting information for posts | |
| Marisa et al. | Intelligent Gamification Mechanics Using Fuzzy‐AHP and K‐Means to Provide Matched Partner Reference |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190308 |
|
| WD01 | Invention patent application deemed withdrawn after publication |