CN109933656A - Public sentiment polarity prediction technique, device, computer equipment and storage medium - Google Patents
Public sentiment polarity prediction technique, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109933656A CN109933656A CN201910199451.5A CN201910199451A CN109933656A CN 109933656 A CN109933656 A CN 109933656A CN 201910199451 A CN201910199451 A CN 201910199451A CN 109933656 A CN109933656 A CN 109933656A
- Authority
- CN
- China
- Prior art keywords
- public sentiment
- characteristic
- data
- affective characteristics
- model
- 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.)
- Granted
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to public sentiment polarity prediction technique, device, computer equipment and storage medium, this method includes obtaining public sentiment data；AC automatic machine based on even numbers group dictionary tree carries out affective characteristics information extraction to data to be analyzed, to obtain characteristic；Polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；Export the prediction result.The present invention constructs sentiment dictionary by the storage organization of even numbers group dictionary tree, reduce the amount of physical memory of disk IO read-write number and occupancy, public sentiment data is subjected to affective characteristics information extraction in sentiment dictionary using the AC automatic machine based on even numbers group dictionary tree, state transfer is converted by charactor comparison, completely without backtracking when scanning data to be analyzed, avoid multiple rollback scanning problem, polarity prediction is carried out to characteristic by public sentiment polarity prediction model, effectively improves the efficiency and accuracy of public sentiment polarity forecast analysis.
Description
Technical field
The present invention relates to information processing methods, more specifically refer to public sentiment polarity prediction technique, device, computer equipment
And storage medium.
Background technique
With the fast development of the applications such as wechat, microblogging, more and more netizens express viewpoint by internet.Network
The fusion of information and social information to society generate influence it is increasing, or even be related to country information security and Changzhi it is long
Peace.Since the information content on internet is very huge, the public sentiment data of magnanimity can not be handled, by artificial method to complete
Face, complete acquisition public sentiment totality situation situation, need automatically to supervise public feelings information by feeling polarities analytical technology
Control and analysis.
Existing the analysis of public opinion application system, what is generallyd use is keyword analysis method and keyword analysis, not only low efficiency, accuracy rate
It is not high.Based on traditional Chinese word segmentation, carrying out pattern match will repeatedly retract scan text, and effectiveness of performance is relatively low；It is existing
The analysis of public opinion application system is using more cursorily statistical method calculates feeling polarities, due to the limitation and context of characteristic information
The influence of context, accuracy rate be not high；Public sentiment sentiment dictionary occupancy memory space is bigger, brings the loss in performance.
Therefore, it is necessary to design a kind of new method, the speed to solve Chinese word segmentation is low, polarity predictablity rate is low,
Big problem is lost in performance.
Summary of the invention
It is an object of the invention to overcome the deficiencies of existing technologies, public sentiment polarity prediction technique, device, computer are provided and set
Standby and storage medium.
To achieve the above object, the invention adopts the following technical scheme: public sentiment polarity prediction technique, comprising:
Obtain public sentiment data；
AC automatic machine based on even numbers group dictionary tree carries out affective characteristics information extraction to data to be analyzed, to obtain feature
Data；
Polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；
Export the prediction result.
Its further technical solution are as follows: the AC automatic machine based on even numbers group dictionary tree is treated point based on sentiment dictionary
The multimode matching algorithm that data carry out affective characteristics information extraction is analysed, the sentiment dictionary is constructed based on even numbers group dictionary tree
's.
Its further technical solution are as follows: the AC automatic machine based on even numbers group dictionary tree carries out emotion to data to be analyzed
Feature information extraction, to obtain characteristic, comprising:
Pattern match is carried out to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree, to obtain output knot
Fruit；
Affective characteristics information extraction is carried out to output result, to obtain characteristic.
Its further technical solution are as follows: the described pair of AC automatic machine based on even numbers group dictionary tree carries out pattern match, with
To output result, comprising:
Several characters are split as to the data to be analyzed；
According to the chracter search sentiment dictionary；
Judge whether the character matches；
If matching, matched character is exported into setting set, to form output result；
Judge whether current character is last character；
If so, affective characteristics information extraction is carried out into described pair of output result, to obtain characteristic；
If it is not, then obtaining next character；
It returns described according to the chracter search sentiment dictionary；
If mismatching, the character that failure function is directed toward is turned to；
Whether the character for judging that the failure function is directed toward is empty；
If it is not, then exporting in the character extremely setting set that the failure function is directed toward, to form output result；
It returns and described judges whether current character is last character；
If so, into end step.
Its further technical solution are as follows: described pair of output result carries out affective characteristics information extraction, to obtain characteristic,
Include:
Output result is divided into several atom words；
Establish the adjacency list for storing array figure；
The position of atom word is determined using the offset of atom word；
Atom word is added to the corresponding position of the array in adjacency list；
The distance between the atom word of two nodes in array is calculated based on viterbi algorithm；
It gives a mark to the entire array figure of adjacency list storage；
By it is described apart from shortest atom word, position and attribute information be added setting affective characteristics data acquisition system,
To form characteristic.
Its further technical solution are as follows: it is described that polarity prediction is carried out to characteristic by public sentiment polarity prediction model, with
It obtains in prediction result, the public sentiment polarity prediction model is by the extracted affective characteristics data set input of sentiment dictionary
After obtaining characteristic of division in XGBoost model, characteristic of division is input to the resulting model of Logic Regression Models training.
Its further technical solution are as follows: the public sentiment polarity prediction model is by the extracted affective characteristics of sentiment dictionary
After obtaining characteristic of division in data set input XGBoost model, characteristic of division is input to Logic Regression Models and is trained institute
The model obtained, comprising:
According to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
Decision tree is input in XGBoost model, it is special to obtain XGBoost model and the extracted emotion of sentiment dictionary
Levy the residual error of data set reality output；
New decision tree is constructed according to the residual error；
Using decision tree described in new decision tree iteration, to obtain the combination of affective characteristics information；
It combines the affective characteristics information in input logic regression model, Logic Regression Models is trained；
The processing of model persistence is carried out to the Logic Regression Models after training, to obtain public sentiment polarity prediction model.
The present invention also provides public sentiment polarity prediction meanss, comprising:
Public sentiment data acquiring unit, for obtaining public sentiment data；
Extraction unit carries out affective characteristics information to data to be analyzed for the AC automatic machine based on even numbers group dictionary tree and mentions
It takes, to obtain characteristic；
Predicting unit, for carrying out polarity prediction to characteristic by public sentiment polarity prediction model, to obtain prediction knot
Fruit；
Output unit, for exporting the prediction result.
The present invention also provides a kind of computer equipment, the computer equipment includes memory and processor, described to deposit
Computer program is stored on reservoir, the processor realizes above-mentioned method when executing the computer program.
The present invention also provides a kind of storage medium, the storage medium is stored with computer program, the computer journey
Sequence can realize above-mentioned method when being executed by processor.
Compared with the prior art, the invention has the advantages that: the present invention is by the storage organization of even numbers group dictionary tree come structure
Sentiment dictionary is built, the amount of physical memory of disk IO read-write number and occupancy is reduced, utilizes the AC based on even numbers group dictionary tree
Public sentiment data is carried out affective characteristics information extraction by automatic machine in sentiment dictionary, is converted state transfer for charactor comparison, is swept
Completely without backtracking when retouching data to be analyzed, multiple rollback scanning problem is avoided, by public sentiment polarity prediction model to spy
It levies data and carries out polarity prediction, effectively improve the efficiency and accuracy of public sentiment polarity forecast analysis.
The invention will be further described in the following with reference to the drawings and specific embodiments.
Detailed description of the invention
Technical solution in order to illustrate the embodiments of the present invention more clearly, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, drawings in the following description are some embodiments of the invention, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the application scenarios schematic diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 2 is the flow diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 3 is the sub-process schematic diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 4 is the sub-process schematic diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 5 is the sub-process schematic diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 6 is the sub-process schematic diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention；
Fig. 7 is state transition diagram provided in an embodiment of the present invention；
Fig. 8 is the schematic diagram of failure function provided in an embodiment of the present invention；
Fig. 9 is that public sentiment polarity provided in an embodiment of the present invention predicts schematic diagram；
Figure 10 is the schematic block diagram of public sentiment polarity prediction meanss provided in an embodiment of the present invention；
Figure 11 is the schematic block diagram of computer equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
Fig. 1 and Fig. 2 are please referred to, Fig. 1 is that the application scenarios of public sentiment polarity prediction technique provided in an embodiment of the present invention are illustrated
Figure.Fig. 2 is the schematic flow chart of public sentiment polarity prediction technique provided in an embodiment of the present invention.The public sentiment polarity prediction technique is answered
For in server.Server uses pretreatment operation to it, based on even numbers group word according to the target public sentiment web site contents crawled
The AC automatic machine of allusion quotation tree is analyzed and the prediction of public sentiment polarity prediction model, to obtain public sentiment polarity results, and exports to terminal
Display.
Fig. 2 is the flow diagram of public sentiment polarity prediction technique provided in an embodiment of the present invention.As shown in Fig. 2, this method
Include the following steps S110 to S130.
S110, public sentiment data is obtained.
In the present embodiment, public sentiment data refers to the data for representing commentator's emotion.
In one embodiment, above-mentioned step S110 can comprise the following steps that
Crawl target public sentiment web site contents；
In the present embodiment, target public sentiment web site contents refer to the content from webpage and website.It is climbed using crawler technology
Take target public sentiment web site contents.
The target public sentiment web site contents are pre-processed, web page analysis and denoising, to obtain public sentiment data.
In the present embodiment, it needs to carry out preliminary treatment to target public sentiment web site contents, obtains public sentiment data, removal need not
The data wanted.
S120, the AC automatic machine based on even numbers group dictionary tree carry out affective characteristics information extraction to data to be analyzed, with
To characteristic.
In the present embodiment, the AC automatic machine based on even numbers group dictionary tree is to be carried out based on sentiment dictionary to data to be analyzed
The multimode matching algorithm of affective characteristics information extraction.
The sentiment dictionary is constructed based on even numbers group dictionary tree.
In the present embodiment, sentiment dictionary refers to the set of all word compositions with emotional color.
Based on the dictionaries store structure of even numbers group dictionary tree, first determines the state of word and turn to function, and calculate mistake
Function is imitated, the calculating of output function is then to be interspersed among two steps to complete, and even numbers group dictionary tree is the dictionary tree of a compression, is led to
Crossing using two one-dimension array BASE and CHECK indicates entirely to set.
For example, the sentiment dictionary being made of { compatriots team, compatriots Chinese national team } is constructed, is turned to construct
Function needs to construct a state transition diagram.Firstly, state transition diagram only includes an initial state 0, by addition one from
The mode in the path that initial state is set out, successively inputs each keyword p into figure, and new vertex and side are added into chart
In, it is final to generate the path that spell out keyword p, in order to complete to turn to the building of function, in addition to bebinning character
Other each characters, all increase by one from state 0 to the circulation of state 0, with the state transition diagram being illustrated in fig. 7 shown below,
This figure, which just represents, turns to function.
Failure function is first to calculate the failure functional value for the state that all depth are 1 according to function foundation is turned to, calculate institute
Having depth is 2 state, and so on, until all failure functional values in addition to the state of state 0 are all calculated, state 0
Depth do not define, obtain i=1,2,3,4,5,6,7,8,9 when, corresponding state value was 0,0,0,1,2,0,3,0,3；Finally
Obtain failure function as described in Figure 8.
In addition, needing for sentiment dictionary to be loaded into memory, using one-piece design mould when first time running AC automatic machine
Formula is designed the model object of the sentiment dictionary of AC automatic machine, and the model after persistence is loaded when running first time
To memory, behind called every time there is no need to execute the operation such as compiling and load again, realize primary compiling load, be run multiple times,
The high efficiency feature of internal storage access is made full use of, the efficiency of affective characteristics information extraction is improved.It is compressed using even numbers group dictionary tree
Memory space compresses the memory space for reducing disk IO read-write number and occupancy, using storage to improve the efficiency of internal storage access.
Characteristic refers to the data with affective characteristics information, that is, represents the word of commentator's emotion.
In one embodiment, referring to Fig. 3, above-mentioned step S120 may include step S121~S122.
S121, pattern match is carried out to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree, it is defeated to obtain
Result out；
Output result refers to the set of words to match with emotion word.
In one embodiment, referring to Fig. 4, above-mentioned step S121 may include step S121a~S121i.
S121a, several characters are split as to the data to be analyzed；
S121b, according to the chracter search sentiment dictionary.
The searching character in sentiment dictionary, due to sentiment dictionary be by steering function and failure function it is built-up,
When AC automatic machine carries out affective characteristics information extraction, state transfer dexterously is converted by charactor comparison, to carry out character and feelings
The matching treatment of sense dictionary avoids multiple rollback scanning problem completely without backtracking when scanning data to be analyzed.
S121c, judge whether the character matches；
If S121d, matching, matched character is exported into setting set, to form output result.
When character match, when the output function of sentiment dictionary is not sky, AC automatic machine is output match pattern, output matching
Character to setting set in, to form output result.
S121e, judge whether current character is last character；
If so, entering step S122；
S121f, if it is not, then obtaining next character；
Return to the step S121b；
If S121g, mismatching, the character that failure function is directed toward is turned to.
When current character mismatches, then show that current character fails, then AC automatic machine turns to failure function and refers to
To character.
Whether S121h, the character for judging that the failure function is directed toward are empty；
S121i, it is extremely set in set if it is not, then exporting the character that the failure function is directed toward, to form output result.
When the character that failure function is directed toward be not it is empty, then the character is exported into setting set, to form output result.
Return to the step S121e；
If so, into end step.
Above-mentioned step is recycled, all characters in data to be analyzed are matched, to obtain completely exporting result.
S122, affective characteristics information extraction is carried out to output result, to obtain characteristic.
A word is provided in affectional priori knowledge by sentiment dictionary, indicates the word under most of contexts
Feeling polarities and its information such as intensity.Affective characteristics information is extracted based on sentiment dictionary, extracting has value in public sentiment text
Emotion information, will have not a particle of rule non-structured text be converted into computer it will be appreciated that identification structured features letter
Breath.Finally obtained affective characteristics information, that is, characteristic presentation format: { emotion word, part of speech, position, Sentiment orientation, feelings in sentence
Feel intensity }.
In one embodiment, referring to Fig. 5, above-mentioned step S122 may include step S1221~S1227.
S1221, output result is divided into several atom words.
Atom word refers to the word of minimum unit.It is realized based on AC automatic machine and a sentence is splitted into all possible original
Sub- word.
S1222, adjacency list for storing array figure is established.
Figure is stored using an adjacency list.
S1223, the position that atom word is determined using the offset of atom word；
S1224, atom word is added to the corresponding position of the array in adjacency list；
Judged using the offset offset of each atom word term it where, atom word term is added
Enter at adjacency list array terms [offset].
S1225, the distance between atom word word frequency of two nodes in array is calculated based on viterbi algorithm；
S1226, it gives a mark to the entire array figure of adjacency list storage；
The distance between the atom word term of two nodes is calculated based on viterbi algorithm, is assigned with one for each node
A distance represents the length of the accumulative shortest path from root node to present node, is then entirely schemed by depth-first traversal
It gives a mark, as long as marking is plus the distance from root node to present node every time.
S1227, by it is described apart from shortest atom word, position and attribute information be added setting affective characteristics data
Set, to form characteristic.
By on shortest path emotion word and the information such as position and attribute be added to affective characteristics data acquisition system.In this reality
It applies in example, attribute information refers to part of speech, the information such as position, Sentiment orientation, emotional intensity in sentence.
S130, polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；
In the present embodiment, prediction result refers to the polarity number of public sentiment data.The public sentiment polarity prediction model is to pass through
After obtaining characteristic of division in the extracted affective characteristics data set input XGBoost model of sentiment dictionary, characteristic of division is inputted
Resulting model is trained to Logic Regression Models.
Input feature vector data utilize XGBoost Construction of A Model new feature, and the new feature vector of construction is value 0/1, to
Each element of amount corresponds to the leaf node set in XGBoost model.When a sample point finally falls in this by certain tree
Tree a leaf node on, then in new feature vector the corresponding element value of this leaf node be 1, and this tree
The corresponding element value of other leaf nodes is 0, and the length of new feature vector is equal to the leaf that all trees include in XGBoost model
Finally original feature training pattern together is added, to obtain public sentiment polarity prediction model in these new features by the sum of nodal point number.Often
A output individually set is considered as the classification input feature vector of sparse linear classifier, as shown in figure 9, input division has two trees,
Upper tree is there are two leaf node, and there are three leaf nodes for lower tree, and final feature is the vector of five dimensions.For inputting x, upper tree
Second node then encodes [0,1], it is assumed that he falls in first node of lower tree, encodes [1,0,0], falls in the coding so final
For [0,1,1,0,0], the input feature vector as prediction model will be encoded, be input in Logic Regression Models and predicted.
In one embodiment, referring to Fig. 6, above-mentioned public sentiment polarity prediction model is by the extracted feelings of sentiment dictionary
After obtaining characteristic of division in sense characteristic data set input XGBoost model, characteristic of division is input to Logic Regression Models and is carried out
The resulting model of training, including step S131~S136.
S131, according to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
S132, decision tree is input in XGBoost model, it is extracted to obtain XGBoost model and sentiment dictionary
The residual error of affective characteristics data set reality output.
S133, new decision tree is constructed according to the residual error；
S134, using decision tree described in new decision tree iteration, to obtain the combination of affective characteristics information.
Above-mentioned XGBoost (extreme gradient is promoted, eXtreme Gradient Boosting) model is large-scale parallel
The tool of boosted tree, it is current most fast best open source boosted tree kit, and Xgboost model is very much
CART regression tree is integrated.
A decision tree is reconstructed in the residual error that existing model and actual sample export, is constantly iterated.Often
An iteration can all generate the biggish characteristic of division of gain, obtain multiple affective characteristics with discrimination by more trees
Information combination.
S135, it combines the affective characteristics information in input logic regression model, Logic Regression Models is trained；
S136, the processing of model persistence is carried out to the Logic Regression Models after training, to obtain public sentiment polarity prediction model.
The affective characteristics information combine the input as Logic Regression Models；Training Logic Regression Models simultaneously hold model
Longization.
XGBoost is the efficient realization of GBDT algorithm, supports parallel processing, and base learner uses CART regression tree, canonical
The leaf node quantity for changing Xiang Yushu is related with the value of leaf node；XGBoost according to Taylor expansion come approximate objective function,
It calculates puppet residual error learning function FM (x), not only used first derivative, also use second dervative, while model cost function
In be also added into regular terms, for the complexity of Controlling model, so that the model that study comes out is simpler.
Network public-opinion content of text is predicted using public sentiment polarity prediction model to obtain polarity results, and uses F-
Score evaluates final classification result, is defined as follows:
F-Score=(2 × Precision × Recall)/(Precision+Recall), wherein Precision is represented
Accuracy rate, Recall represent recall rate.
The example number that Precision=class is correctly classified/public sentiment polarity prediction model predicts the total of certain class example
Number
The sum of certain class example in example number/test data that Recall=class is correctly classified.
S140, the output prediction result.
Prediction result output uses json format string, and output format example is as follows: { " sentiTrend ": " just
Face ", " sentineg ": 0.278, " sentipos ": 0.722 }.
It is tested using the microblog data 20w item of crawler capturing, different public sentiment polarity prediction algorithm accuracy rate comparative situations are such as
Shown in Tables 1 and 2.
The comparison of 1. characteristic extraction rate of table
| Algorithm | Dictionary scale | Extraction rate |
| IK participle | 35w | 80w/s |
| Ansj participle | 35w | 210w/s |
| Fnlp participle | 35w | 120w/s |
| Even numbers group AC automatic machine | 35w | 1600w/s |
The comparison of 2. accuracy rate of table
| Prediction algorithm | Accuracy rate | F1 |
| Keyword statistical method | 0.703 | 0.633 |
| Logistics algorithm | 0.718 | 0.646 |
| GBDT+lr algorithm | 0.803 | 0.725 |
| XGBoost+lr algorithm | 0.812 | 0.736 |
Above-mentioned public sentiment polarity prediction technique, constructs sentiment dictionary by the storage organization of even numbers group dictionary tree, reduces
The amount of physical memory of disk IO read-write number and occupancy, using AC automatic machine based on even numbers group dictionary tree by public sentiment data
Affective characteristics information extraction is carried out in sentiment dictionary, is converted state transfer for charactor comparison, is scanned complete when data to be analyzed
It does not need to recall entirely, avoids multiple rollback scanning problem, it is pre- to carry out polarity to characteristic by public sentiment polarity prediction model
It surveys, effectively improves the efficiency and accuracy of public sentiment polarity forecast analysis.
Figure 10 is a kind of schematic block diagram of public sentiment polarity prediction meanss provided in an embodiment of the present invention.As shown in Figure 10,
Corresponding to the above public sentiment polarity prediction technique, the present invention also provides a kind of public sentiment polarity prediction meanss.Public sentiment polarity prediction dress
It sets including the unit for executing above-mentioned public sentiment polarity prediction technique, which can be configured in server.
Specifically, referring to Fig. 10, the public sentiment polarity prediction meanss include:
Public sentiment data acquiring unit 301, for obtaining public sentiment data；
Extraction unit 302 carries out affective characteristics letter to data to be analyzed for the AC automatic machine based on even numbers group dictionary tree
Breath extracts, to obtain characteristic；
Predicting unit 303, for carrying out polarity prediction to characteristic by public sentiment polarity prediction model, to be predicted
As a result；
Output unit 304, for exporting the prediction result.
In one embodiment, the extraction unit 302 includes:
Coupling subelement, for carrying out mode to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree
Match, to obtain output result；
Characteristic forms subelement, for carrying out affective characteristics information extraction to output result, to obtain characteristic.
In one embodiment, above-mentioned coupling subelement includes:
Module is split, for being split as several characters to the data to be analyzed；
Search module, for according to the chracter search sentiment dictionary；
Character judgement module, for judging whether the character matches；
First output module, if matched character is exported into setting set, for matching to form output result；
Last character judgment module, for judging whether current character is last character；If so, into described right
It exports result and carries out affective characteristics information extraction, to obtain characteristic；
Character obtains module, for if it is not, then obtaining next character；It returns described according to the chracter search emotion word
Allusion quotation；
Steering module, if turning to the character that failure function is directed toward for mismatching；
It is directed toward judgment module, whether the character for judging that the failure function is directed toward is empty；If so, entering terminates step
Suddenly；
Second output module, for if it is not, then exporting in the character extremely setting set that the failure function is directed toward, to be formed
Export result；It returns and described judges whether current character is last character.
In one embodiment, above-mentioned characteristic formation subelement includes:
Division module is divided into several atom words for that will export result；
Adjacency list establishes module, for establishing the adjacency list for storing array figure；
Position determination module, for determining the position of atom word using the offset of atom word；
Module is added, for atom word to be added to the corresponding position of the array in adjacency list；
Distance calculation module, for based on viterbi algorithm calculate array in two nodes atom word between away from
From；
Scoring modules, for giving a mark to the entire array figure that adjacency list stores；
Module is integrated, for the emotion set to be added apart from shortest atom word, position and attribute information by described
Characteristic set, to form characteristic.
In one embodiment, above-mentioned device further include:
Model training unit, for being inputted in XGBoost model by the extracted affective characteristics data set of sentiment dictionary
After obtaining characteristic of division, characteristic of division is input to Logic Regression Models and is trained, to obtain public sentiment polarity prediction model.
In one embodiment, above-mentioned model training unit includes:
First construction subelement, for according to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
First input subelement, for decision tree to be input in XGBoost model, to obtain XGBoost model and feelings
Feel the residual error of the extracted affective characteristics data set reality output of dictionary；
Second construction subelement, for constructing new decision tree according to the residual error；
Iteration subelement, for being combined with obtaining affective characteristics information using decision tree described in new decision tree iteration；
Combination input subelement returns logic for combining the affective characteristics information in input logic regression model
Model is returned to be trained；
Subelement is handled, for carrying out the processing of model persistence to the Logic Regression Models after training, to obtain public sentiment pole
Property prediction model.
It should be noted that it is apparent to those skilled in the art that, above-mentioned public sentiment polarity prediction meanss
It, can be for convenience of description and simple with reference to the corresponding description in preceding method embodiment with the specific implementation process of each unit
Clean, details are not described herein.
Above-mentioned public sentiment polarity prediction meanss can be implemented as a kind of form of computer program, which can be
It is run in computer equipment as shown in figure 11.
Figure 11 is please referred to, Figure 11 is a kind of schematic block diagram of computer equipment provided by the embodiments of the present application.The calculating
Machine equipment 500 is server.
Refering to fig. 11, which includes processor 502, memory and the net connected by system bus 501
Network interface 505, wherein memory may include non-volatile memory medium 503 and built-in storage 504.
The non-volatile memory medium 503 can storage program area 5031 and computer program 5032.The computer program
5032 include program instruction, which is performed, and processor 502 may make to execute a kind of public sentiment polarity prediction technique.
The processor 502 is for providing calculating and control ability, to support the operation of entire computer equipment 500.
The built-in storage 504 provides environment for the operation of the computer program 5032 in non-volatile memory medium 503, should
When computer program 5032 is executed by processor 502, processor 502 may make to execute a kind of public sentiment polarity prediction technique.
The network interface 505 is used to carry out network communication with other equipment.It will be understood by those skilled in the art that in Figure 11
The structure shown, only the block diagram of part-structure relevant to application scheme, does not constitute and is applied to application scheme
The restriction of computer equipment 500 thereon, specific computer equipment 500 may include more more or fewer than as shown in the figure
Component perhaps combines certain components or with different component layouts.
Wherein, the processor 502 is for running computer program 5032 stored in memory, to realize following step
It is rapid:
Obtain public sentiment data；
AC automatic machine based on even numbers group dictionary tree carries out affective characteristics information extraction to data to be analyzed, to obtain feature
Data；
Polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；
Export the prediction result.
Wherein, the AC automatic machine based on even numbers group dictionary tree is to carry out emotion to data to be analyzed based on sentiment dictionary
The multimode matching algorithm of feature information extraction, the sentiment dictionary are constructed based on even numbers group dictionary tree.
In one embodiment, processor 502 is realizing the AC automatic machine based on even numbers group dictionary tree to number to be analyzed
According to affective characteristics information extraction is carried out, when obtaining characteristic data step, it is implemented as follows step:
Pattern match is carried out to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree, to obtain output knot
Fruit；
Affective characteristics information extraction is carried out to output result, to obtain characteristic.
In one embodiment, processor 502 is realizing the described pair of AC automatic machine progress mode based on even numbers group dictionary tree
Matching is implemented as follows step when obtaining output result step:
Several characters are split as to the data to be analyzed；
According to the chracter search sentiment dictionary；
Judge whether the character matches；
If matching, matched character is exported into setting set, to form output result；
Judge whether current character is last character；
If so, affective characteristics information extraction is carried out into described pair of output result, to obtain characteristic；
If it is not, then obtaining next character；
It returns described according to the chracter search sentiment dictionary；
If mismatching, the character that failure function is directed toward is turned to；
Whether the character for judging that the failure function is directed toward is empty；
If it is not, then exporting in the character extremely setting set that the failure function is directed toward, to form output result；
It returns and described judges whether current character is last character；
If so, into end step.
Wherein, described that polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain in prediction result,
The public sentiment polarity prediction model is to input in XGBoost model to obtain by the extracted affective characteristics data set of sentiment dictionary
After characteristic of division, characteristic of division is input to the resulting model of Logic Regression Models training.
In one embodiment, processor 502 is realizing that the public sentiment polarity prediction model is extracted by sentiment dictionary
Affective characteristics data set input XGBoost model in obtain characteristic of division after, characteristic of division is input to Logic Regression Models
When being trained resulting model step, it is implemented as follows step:
According to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
Decision tree is input in XGBoost model, it is special to obtain XGBoost model and the extracted emotion of sentiment dictionary
Levy the residual error of data set reality output；
New decision tree is constructed according to the residual error；
Using decision tree described in new decision tree iteration, to obtain the combination of affective characteristics information；
It combines the affective characteristics information in input logic regression model, Logic Regression Models is trained；
The processing of model persistence is carried out to the Logic Regression Models after training, to obtain public sentiment polarity prediction model.
It should be appreciated that in the embodiment of the present application, processor 502 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..Wherein, general processor can be microprocessor or
Person's processor is also possible to any conventional processor etc..
Those of ordinary skill in the art will appreciate that be realize above-described embodiment method in all or part of the process,
It is that relevant hardware can be instructed to complete by computer program.The computer program includes program instruction, computer journey
Sequence can be stored in a storage medium, which is computer readable storage medium.The program instruction is by the department of computer science
At least one processor in system executes, to realize the process step of the embodiment of the above method.
Therefore, the present invention also provides a kind of storage mediums.The storage medium can be computer readable storage medium.This is deposited
Storage media is stored with computer program, and processor is made to execute following steps when wherein the computer program is executed by processor:
Obtain public sentiment data；
AC automatic machine based on even numbers group dictionary tree carries out affective characteristics information extraction to data to be analyzed, to obtain feature
Data；
Polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；
Export the prediction result.
Wherein, the AC automatic machine based on even numbers group dictionary tree is to carry out emotion to data to be analyzed based on sentiment dictionary
The multimode matching algorithm of feature information extraction, the sentiment dictionary are constructed based on even numbers group dictionary tree.
In one embodiment, the processor is realized described based on even numbers group dictionary tree in the execution computer program
AC automatic machine to data to be analyzed carry out affective characteristics information extraction, when obtaining characteristic data step, be implemented as follows
Step:
Pattern match is carried out to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree, to obtain output knot
Fruit；
Affective characteristics information extraction is carried out to output result, to obtain characteristic.
In one embodiment, the processor realizes that described pair is based on even numbers group dictionary executing the computer program
The AC automatic machine of tree carries out pattern match, when obtaining output result step, is implemented as follows step:
Several characters are split as to the data to be analyzed；
According to the chracter search sentiment dictionary；
Judge whether the character matches；
If matching, matched character is exported into setting set, to form output result；
Judge whether current character is last character；
If so, affective characteristics information extraction is carried out into described pair of output result, to obtain characteristic；
If it is not, then obtaining next character；
It returns described according to the chracter search sentiment dictionary；
If mismatching, the character that failure function is directed toward is turned to；
Whether the character for judging that the failure function is directed toward is empty；
If it is not, then exporting in the character extremely setting set that the failure function is directed toward, to form output result；
It returns and described judges whether current character is last character；
If so, into end step.
In one embodiment, the processor realizes described pair of output result progress feelings executing the computer program
Feel feature information extraction, when obtaining characteristic data step, be implemented as follows step:
Output result is divided into several atom words；
Establish the adjacency list for storing array figure；
The position of atom word is determined using the offset of atom word；
Atom word is added to the corresponding position of the array in adjacency list；
The distance between the atom word of two nodes in array is calculated based on viterbi algorithm；
It gives a mark to the entire array figure of adjacency list storage；
By it is described apart from shortest atom word, position and attribute information be added setting affective characteristics data acquisition system,
To form characteristic.
Wherein, described that polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain in prediction result,
The public sentiment polarity prediction model is to input in XGBoost model to obtain by the extracted affective characteristics data set of sentiment dictionary
After characteristic of division, characteristic of division is input to the resulting model of Logic Regression Models training.
In one embodiment, the processor realizes the public sentiment polarity prediction model executing the computer program
It is after being inputted in XGBoost model by the extracted affective characteristics data set of sentiment dictionary and obtaining characteristic of division, classification is special
When sign is input to Logic Regression Models and is trained resulting model step, it is implemented as follows step:
According to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
Decision tree is input in XGBoost model, it is special to obtain XGBoost model and the extracted emotion of sentiment dictionary
Levy the residual error of data set reality output；
New decision tree is constructed according to the residual error；
Using decision tree described in new decision tree iteration, to obtain the combination of affective characteristics information；
It combines the affective characteristics information in input logic regression model, Logic Regression Models is trained；
The processing of model persistence is carried out to the Logic Regression Models after training, to obtain public sentiment polarity prediction model.
The storage medium can be USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), magnetic disk
Or the various computer readable storage mediums that can store program code such as CD.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary.For example, the division of each unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation.Such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.
The steps in the embodiment of the present invention can be sequentially adjusted, merged and deleted according to actual needs.This hair
Unit in bright embodiment device can be combined, divided and deleted according to actual needs.In addition, in each implementation of the present invention
Each functional unit in example can integrate in one processing unit, is also possible to each unit and physically exists alone, can also be with
It is that two or more units are integrated in one unit.
If the integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product,
It can store in one storage medium.Based on this understanding, technical solution of the present invention is substantially in other words to existing skill
The all or part of part or the technical solution that art contributes can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, terminal or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. public sentiment polarity prediction technique characterized by comprising
Obtain public sentiment data；
AC automatic machine based on even numbers group dictionary tree carries out affective characteristics information extraction to data to be analyzed, to obtain characteristic
According to；
Polarity prediction is carried out to characteristic by public sentiment polarity prediction model, to obtain prediction result；
Export the prediction result.
2. public sentiment polarity prediction technique according to claim 1, which is characterized in that the AC based on even numbers group dictionary tree
Automatic machine is the multimode matching algorithm that based on sentiment dictionary data to be analyzed are carried out with affective characteristics information extraction, the emotion word
Allusion quotation is constructed based on even numbers group dictionary tree.
3. public sentiment polarity prediction technique according to claim 2, which is characterized in that the AC based on even numbers group dictionary tree
Automatic machine carries out affective characteristics information extraction to data to be analyzed, to obtain characteristic, comprising:
Pattern match is carried out to data to be analyzed using the AC automatic machine based on even numbers group dictionary tree, to obtain output result；
Affective characteristics information extraction is carried out to output result, to obtain characteristic.
4. public sentiment polarity prediction technique according to claim 3, which is characterized in that described pair based on even numbers group dictionary tree
AC automatic machine carries out pattern match, to obtain output result, comprising:
Several characters are split as to the data to be analyzed；
According to the chracter search sentiment dictionary；
Judge whether the character matches；
If matching, matched character is exported into setting set, to form output result；
Judge whether current character is last character；
If so, affective characteristics information extraction is carried out into described pair of output result, to obtain characteristic；
If it is not, then obtaining next character；
It returns described according to the chracter search sentiment dictionary；
If mismatching, the character that failure function is directed toward is turned to；
Whether the character for judging that the failure function is directed toward is empty；
If it is not, then exporting in the character extremely setting set that the failure function is directed toward, to form output result；
It returns and described judges whether current character is last character；
If so, into end step.
5. public sentiment polarity prediction technique according to claim 4, which is characterized in that it is special that described pair of output result carries out emotion
Information extraction is levied, to obtain characteristic, comprising:
Output result is divided into several atom words；
Establish the adjacency list for storing array figure；
The position of atom word is determined using the offset of atom word；
Atom word is added to the corresponding position of the array in adjacency list；
The distance between the atom word of two nodes in array is calculated based on viterbi algorithm；
It gives a mark to the entire array figure of adjacency list storage；
By the affective characteristics data acquisition system that setting is added apart from shortest atom word, position and attribute information, with shape
At characteristic.
6. public sentiment polarity prediction technique according to claim 2, which is characterized in that described to pass through public sentiment polarity prediction model
Polarity prediction is carried out to characteristic, to obtain in prediction result, the public sentiment polarity prediction model is by sentiment dictionary institute
After obtaining characteristic of division in the affective characteristics data set input XGBoost model of extraction, characteristic of division is input to logistic regression
The resulting model of model training.
7. public sentiment polarity prediction technique according to claim 6, which is characterized in that the public sentiment polarity prediction model is logical
It crosses in the extracted affective characteristics data set input XGBoost model of sentiment dictionary after obtaining characteristic of division, characteristic of division is defeated
Enter to Logic Regression Models and be trained resulting model, comprising:
According to the extracted affective characteristics dataset construction decision tree of sentiment dictionary；
Decision tree is input in XGBoost model, to obtain XGBoost model and the extracted affective characteristics number of sentiment dictionary
According to the residual error of collection reality output；
New decision tree is constructed according to the residual error；
Using decision tree described in new decision tree iteration, to obtain the combination of affective characteristics information；
It combines the affective characteristics information in input logic regression model, Logic Regression Models is trained；
The processing of model persistence is carried out to the Logic Regression Models after training, to obtain public sentiment polarity prediction model.
8. public sentiment polarity prediction meanss characterized by comprising
Public sentiment data acquiring unit, for obtaining public sentiment data；
Extraction unit carries out affective characteristics information extraction to data to be analyzed for the AC automatic machine based on even numbers group dictionary tree,
To obtain characteristic；
Predicting unit, for carrying out polarity prediction to characteristic by public sentiment polarity prediction model, to obtain prediction result；
Output unit, for exporting the prediction result.
9. a kind of computer equipment, which is characterized in that the computer equipment includes memory and processor, on the memory
It is stored with computer program, the processor is realized as described in any one of claims 1 to 7 when executing the computer program
Method.
10. a kind of storage medium, which is characterized in that the storage medium is stored with computer program, the computer program quilt
Processor can realize the method as described in any one of claims 1 to 7 when executing.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910199451.5A CN109933656B (en) | 2019-03-15 | 2019-03-15 | Public opinion polarity prediction method, public opinion polarity prediction device, computer equipment and storage medium |
| PCT/CN2019/089224 WO2020186627A1 (en) | 2019-03-15 | 2019-05-30 | Public opinion polarity prediction method and apparatus, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910199451.5A CN109933656B (en) | 2019-03-15 | 2019-03-15 | Public opinion polarity prediction method, public opinion polarity prediction device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109933656A true CN109933656A (en) | 2019-06-25 |
| CN109933656B CN109933656B (en) | 2023-08-15 |
Family
ID=66987288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910199451.5A Active CN109933656B (en) | 2019-03-15 | 2019-03-15 | Public opinion polarity prediction method, public opinion polarity prediction device, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109933656B (en) |
| WO (1) | WO2020186627A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110362669A (en) * | 2019-07-18 | 2019-10-22 | 中科信息安全共性技术国家工程研究中心有限公司 | A kind of algorithm suitable for keyword quick-searching |
| CN110674297A (en) * | 2019-09-24 | 2020-01-10 | 支付宝(杭州)信息技术有限公司 | Public opinion text classification model construction method, public opinion text classification device and public opinion text classification equipment |
| CN111831824A (en) * | 2020-07-16 | 2020-10-27 | 民生科技有限责任公司 | Public opinion positive and negative face classification method |
| CN111859074A (en) * | 2020-07-29 | 2020-10-30 | 东北大学 | Internet public opinion information source influence assessment method and system based on deep learning |
| CN113051925A (en) * | 2019-12-26 | 2021-06-29 | 中国移动通信集团有限公司 | Time identification method, device, equipment and computer storage medium |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112328793B (en) * | 2020-11-09 | 2024-07-09 | 北京小米松果电子有限公司 | Comment text data processing method and device and storage medium |
| CN113642881A (en) * | 2021-08-09 | 2021-11-12 | 平安国际智慧城市科技股份有限公司 | Public opinion data risk identification method and device, computer equipment and storage medium |
| CN113643060A (en) * | 2021-08-12 | 2021-11-12 | 工银科技有限公司 | Product price prediction method and device |
| CN114117149B (en) * | 2021-11-25 | 2024-08-02 | 深圳前海微众银行股份有限公司 | Sensitive word filtering method and device and storage medium |
| CN114701870B (en) * | 2022-02-11 | 2024-03-29 | 国能黄骅港务有限责任公司 | Feeding system of dumper and high material level detection method and device thereof |
| CN114861027B (en) * | 2022-04-29 | 2024-06-18 | 深圳市东晟数据有限公司 | Multi-dimensional public opinion recommendation method based on big data and natural language processing |
| CN114722723B (en) * | 2022-04-29 | 2024-06-21 | 湖北工业大学 | Emotion tendency prediction method and equipment based on kernel extreme learning machine optimization |
| CN114897270A (en) * | 2022-06-15 | 2022-08-12 | 青岛文达通科技股份有限公司 | Semantic information fused public opinion propagation quantity prediction method and system |
| CN117407527B (en) * | 2023-10-19 | 2024-09-27 | 中诚力锘(北京)教育咨询集团有限公司 | Education field public opinion big data classification method |
| CN117640259B (en) * | 2024-01-25 | 2024-06-04 | 武汉思普崚技术有限公司 | Script step-by-step detection method and device, electronic equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102779174A (en) * | 2012-06-26 | 2012-11-14 | 北京奇虎科技有限公司 | Public opinion information display system and method |
| CN103365991A (en) * | 2013-07-03 | 2013-10-23 | 深圳市华傲数据技术有限公司 | Method for realizing dictionary memory management of Trie tree based on one-dimensional linear space |
| CN106294326A (en) * | 2016-08-23 | 2017-01-04 | 成都科来软件有限公司 | A kind of news report Sentiment orientation analyzes method |
| CN108021569A (en) * | 2016-11-01 | 2018-05-11 | 中国移动通信有限公司研究院 | The structure of AC automatic machines and Chinese multi-model matching method and relevant apparatus |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102200969A (en) * | 2010-03-25 | 2011-09-28 | 日电（中国）有限公司 | Text sentiment polarity classification system and method based on sentence sequence |
| CN105512687A (en) * | 2015-12-15 | 2016-04-20 | 北京锐安科技有限公司 | Emotion classification model training and textual emotion polarity analysis method and system |
-
2019
- 2019-03-15 CN CN201910199451.5A patent/CN109933656B/en active Active
- 2019-05-30 WO PCT/CN2019/089224 patent/WO2020186627A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102779174A (en) * | 2012-06-26 | 2012-11-14 | 北京奇虎科技有限公司 | Public opinion information display system and method |
| CN103365991A (en) * | 2013-07-03 | 2013-10-23 | 深圳市华傲数据技术有限公司 | Method for realizing dictionary memory management of Trie tree based on one-dimensional linear space |
| CN106294326A (en) * | 2016-08-23 | 2017-01-04 | 成都科来软件有限公司 | A kind of news report Sentiment orientation analyzes method |
| CN108021569A (en) * | 2016-11-01 | 2018-05-11 | 中国移动通信有限公司研究院 | The structure of AC automatic machines and Chinese multi-model matching method and relevant apparatus |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110362669A (en) * | 2019-07-18 | 2019-10-22 | 中科信息安全共性技术国家工程研究中心有限公司 | A kind of algorithm suitable for keyword quick-searching |
| CN110674297A (en) * | 2019-09-24 | 2020-01-10 | 支付宝(杭州)信息技术有限公司 | Public opinion text classification model construction method, public opinion text classification device and public opinion text classification equipment |
| CN110674297B (en) * | 2019-09-24 | 2022-04-29 | 支付宝(杭州)信息技术有限公司 | Public opinion text classification model construction method, public opinion text classification device and public opinion text classification equipment |
| CN113051925A (en) * | 2019-12-26 | 2021-06-29 | 中国移动通信集团有限公司 | Time identification method, device, equipment and computer storage medium |
| CN111831824A (en) * | 2020-07-16 | 2020-10-27 | 民生科技有限责任公司 | Public opinion positive and negative face classification method |
| CN111831824B (en) * | 2020-07-16 | 2024-02-09 | 民生科技有限责任公司 | Public opinion positive and negative surface classification method |
| CN111859074A (en) * | 2020-07-29 | 2020-10-30 | 东北大学 | Internet public opinion information source influence assessment method and system based on deep learning |
| CN111859074B (en) * | 2020-07-29 | 2023-12-29 | 东北大学 | Network public opinion information source influence evaluation method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109933656B (en) | 2023-08-15 |
| WO2020186627A1 (en) | 2020-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109933656A (en) | Public sentiment polarity prediction technique, device, computer equipment and storage medium | |
| CN109241524B (en) | Semantic analysis method and device, computer-readable storage medium and electronic equipment | |
| US20210390700A1 (en) | Referring image segmentation | |
| CN110097085B (en) | Lyric text generation method, training method, device, server and storage medium | |
| CN107085581B (en) | Short text classification method and device | |
| CN112270196B (en) | Entity relationship identification method and device and electronic equipment | |
| CN105117054B (en) | A kind of recognition methods of handwriting input and system | |
| CN109446927B (en) | Double-person interaction behavior identification method based on priori knowledge | |
| CN106407236B (en) | A kind of emotion tendency detection method towards comment data | |
| CN111209384A (en) | Question and answer data processing method and device based on artificial intelligence and electronic equipment | |
| CN110147421B (en) | Target entity linking method, device, equipment and storage medium | |
| CN111125360B (en) | Emotion analysis method and device in game field and model training method and device thereof | |
| KR20210023452A (en) | Apparatus and method for review analysis per attribute | |
| KR20180094664A (en) | Method for information extraction from text data and apparatus therefor | |
| CN110222328B (en) | Method, device and equipment for labeling participles and parts of speech based on neural network and storage medium | |
| CN108108347B (en) | Dialogue mode analysis system and method | |
| CN115062134B (en) | Knowledge question-answering model training and knowledge question-answering method, device and computer equipment | |
| CN110222250A (en) | A kind of emergency event triggering word recognition method towards microblogging | |
| CN108763211A (en) | The automaticabstracting and system of knowledge are contained in fusion | |
| CN115587597B (en) | Sentiment analysis method and device of aspect words based on clause-level relational graph | |
| CN111680529A (en) | Machine translation algorithm and device based on layer aggregation | |
| CN112613321A (en) | Method and system for extracting entity attribute information in text | |
| CN111506726A (en) | Short text clustering method and device based on part-of-speech coding and computer equipment | |
| CN109992665A (en) | A kind of classification method based on the extension of problem target signature | |
| CN112988962B (en) | Text error correction method and device, electronic equipment 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |