CN109460550A - Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data - Google Patents
Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data Download PDFInfo
- Publication number
- CN109460550A CN109460550A CN201811228240.1A CN201811228240A CN109460550A CN 109460550 A CN109460550 A CN 109460550A CN 201811228240 A CN201811228240 A CN 201811228240A CN 109460550 A CN109460550 A CN 109460550A
- Authority
- CN
- China
- Prior art keywords
- security
- report
- analyzed
- participle
- grind
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Human Resources & Organizations (AREA)
- Probability & Statistics with Applications (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Machine Translation (AREA)
Abstract
The present invention discloses a kind of security using big data and grinds report sentiment analysis method, comprising: the security to be analyzed for receiving input grind report；Report is ground to security to be analyzed and carries out subordinate sentence to obtain each subordinate sentence；Each subordinate sentence is segmented to obtain each participle；Prediction class subordinate sentence is determined using sentiment dictionary based on each participle；The affective style respectively segmented in prediction class subordinate sentence is determined according to sentiment dictionary；Affective style based on each participle simultaneously scores to each prediction class subordinate sentence using preset code of points；The whole emotion score that security to be analyzed grind report is calculated based on scoring；The whole emotion score of report is ground according to security to be analyzed and the comparison result of preset fraction threshold value obtains the sentiment analysis result that security to be analyzed grind report.The present invention realizes the sentiment analysis that security are ground with report in the way of big data analysis and intelligent scoring, can solve the problems, such as that the prior art is ground for security reports the efficiency of sentiment analysis scheme and accuracy rate lower, improves efficiency and accuracy rate that security grind report sentiment analysis.
Description
Technical field
The present invention relates to field of computer technology, and in particular to a kind of security using big data grind report sentiment analysis side
Method, device and computer equipment.
Background technique
Security research report can be also simply referred to as security and grind report, refer to related researcher's (such as research in securities broker company
Personnel etc.) on the value of security and Related product or influences the factor of its market price and analyze, made research report
It accuses.
Report is ground to security to analyze, and can be understood security in time and be ground in report about sides such as industry, policy, investment feasibilities
Face situation.For grinding the sentiment analysis of report for security, report is still mainly ground to security by manual type at present and is read
It reads, to analyze the emotion trend of author, but such mode needs to expend a large amount of manpowers, and efficiency and accuracy rate are all lower.This
Outside, also there is the scheme analyzed by sentiment dictionary, for example can be used to judge actively and in terms of Negative Affect using existing
Dictionary resources, such as Hownet Chinese dictionary Hownet, Taiwan Univ. simplified form of Chinese Character feeling polarities dictionary NTUSD, still, Ci Zhongfang
Formula is analyzed for short sentence mostly, and the opposite context for having isolated sentence is difficult to make entire article more perfect
Accurately analysis, especially when for a large amount of sentences with incidence relations such as cause and effect, turnover classes, the accuracy rate of analysis is more paid no attention to
Think.
It is directed to security in the related technology and grinds the efficiency of report sentiment analysis scheme and the problem that accuracy rate is lower, not yet mentions at present
Effective solution scheme out.
Summary of the invention
The purpose of the present invention is to provide a kind of security using big data to grind report sentiment analysis method, apparatus, computer
Equipment and readable storage medium storing program for executing, and then above-mentioned problems of the prior art are overcome to a certain extent, it can be improved to security
Grind the efficiency and accuracy rate of report sentiment analysis.
The present invention is to solve above-mentioned technical problem by following technical proposals:
According to an aspect of the invention, there is provided a kind of security using big data grind report sentiment analysis method, including
Following steps:
S01, the security to be analyzed for receiving input grind report；
S02 grinds report to security to be analyzed and carries out subordinate sentence processing, obtains security to be analyzed and grinds each subordinate sentence in reporting；
S03 grinds each subordinate sentence in report to security to be analyzed and carries out word segmentation processing, obtains security to be analyzed and grinds each point in reporting
Word；
S04 grinds each participle in report based on security to be analyzed, determines security to be analyzed using the sentiment dictionary pre-established
Grind the prediction class subordinate sentence in report；
S05 determines that security to be analyzed grind the emotion class respectively segmented in the prediction class subordinate sentence in report according to the sentiment dictionary
Type；
S06 grinds the affective style respectively segmented in report based on security to be analyzed, using preset code of points to card to be analyzed
Each prediction class subordinate sentence that certificate is ground in report scores；
S07 is ground the scoring of each prediction class subordinate sentence in report based on security to be analyzed, security to be analyzed is calculated and grind report
Whole emotion score；
S08, is analysed to security and grinds the whole emotion score of report to be compared with preset fraction threshold value, and ties according to comparing
Fruit obtains the sentiment analysis result that security to be analyzed grind report.
Further, the sentiment dictionary is established, is included the following steps:
Step 110, report is ground to the security of preset record and carries out subordinate sentence processing, obtained security and grind each first subordinate sentence in reporting；
Step 120, word segmentation processing is carried out to each first subordinate sentence, obtains security and grinds each first participle in reporting；
Step 130, it extracts from all first participles for expressing the participle of positive emotion, for expressing negative emotion
Participle, the participle for expressing negative emotion and the participle for expressing prediction；
Step 140, according to extract result generate respectively positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and
Predict class dictionary；
Step 150, feelings are established based on positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction class dictionary
Feel dictionary.
Further, S04 grinds each participle in report based on security to be analyzed, using the sentiment dictionary that pre-establishes determine to
Analysis security grind the prediction class subordinate sentence in report, comprising:
Each participle that security are ground in report is analysed to be matched with the participle in the sentiment dictionary in prediction class dictionary；
It can be with the participle phase in the sentiment dictionary in prediction class dictionary if security to be analyzed grind any participle in report
Subordinate sentence belonging to any participle is then determined as predicting class subordinate sentence by matching.
Further, S05 determines that security to be analyzed are ground in the prediction class subordinate sentence in report according to the sentiment dictionary and respectively segments
Affective style, comprising:
Be analysed to security grind report in prediction class subordinate sentence in it is each segment respectively with positive emotion dictionary, negative emotion word
Participle in library, negative emotion dictionary is matched；
It can be with positive emotion dictionary, negative emotion dictionary or negative feelings if security to be analyzed grind any participle in report
Participle in sense dictionary matches, then the affective style of any participle is determined as front participle, negative participle or negative
Participle.
Further, S06 grinds the affective style that respectively segments in the prediction class subordinate sentence in report based on security to be analyzed, using pre-
Set code of points to security to be analyzed grind report in each prediction class subordinate sentence score, comprising:
It counts security to be analyzed and grinds front participle in report in each prediction class subordinate sentence, the initial number that negatively segments and no
Surely the quantity segmented；
Whether the quantity for judging negative participle is odd number；
If so, judge odd number negative participle above or below it is adjacent be front participle or negative participle；
If the adjacent front participle in odd number negative participle above or below, the initial number of front participle is subtracted together
The initial number negatively segmented is added one, to segment quantity as front adjusted and negatively segment quantity；If odd number
It is a negative participle above or below it is adjacent be negatively to segment, then by the initial number negatively segmented subtract together by front participle
Initial number adds one, to segment quantity as negative participle quantity adjusted and front；
Be analysed to security grind in each prediction class subordinate sentence in report the difference of front participle quantity and negative participle quantity divided by
Front participle quantity and negative participle quantity and, to obtain the scoring that security to be analyzed grind each prediction class subordinate sentence in reporting.
Further, it includes just dividing and bearing point that security to be analyzed, which grind the scoring of each prediction class subordinate sentence in report, is respectively corresponded
Front prediction class subordinate sentence and negative prediction class subordinate sentence；
S07 grinds the scoring of each prediction class subordinate sentence in report based on security to be analyzed, and security to be analyzed are calculated and grind report
Whole emotion score, comprising:
It calculates separately security to be analyzed and grinds all positive average marks for predicting class subordinate sentences in reporting and all negative prediction classes
The average mark of subordinate sentence；
It is analysed to the average mark and all negative prediction class subordinate sentences of all front prediction class subordinate sentences that security are ground in report
Two times of progress read group totals of average mark, to obtain the whole emotion score that security to be analyzed grind report.
Further, the preset fraction threshold value include the first preset fraction threshold value and the second preset fraction threshold value, first
Preset fraction threshold value is less than the second preset fraction threshold value；
S08 is analysed to security and grinds the whole emotion score of report to be compared with preset fraction threshold value, and ties according to comparing
Fruit obtains the sentiment analysis result that security to be analyzed grind report, comprising:
By whole emotion score lower than the first preset fraction threshold value, between the first preset fraction threshold value and the second preset fraction
Investment securities to be analyzed between threshold value, higher than the second preset fraction threshold value grind the sentiment analysis of report as a result, being identified as seeing
Sky, level watching, see it is more.
To achieve the goals above, the present invention also provides a kind of security using big data to grind report sentiment analysis device, packet
It includes:
Receiving module, security to be analyzed for receiving input grind report；
Subordinate sentence module carries out subordinate sentence processing for grinding report to security to be analyzed, obtains security to be analyzed and grinds each point in reporting
Sentence；
Word segmentation module, each subordinate sentence for being ground in report to security to be analyzed carry out word segmentation processing, obtain security to be analyzed and grind
Each participle in report；
It predicts class subordinate sentence determining module, for grinding each participle in report based on security to be analyzed, utilizes the feelings pre-established
Sense dictionary determines that security to be analyzed grind the prediction class subordinate sentence in report；
Affective style determining module, for determining that security to be analyzed grind the prediction class subordinate sentence in report according to the sentiment dictionary
In the affective style that respectively segments；
Grading module is utilized for being ground the affective style respectively segmented in the prediction class subordinate sentence in report based on security to be analyzed
Preset code of points grinds each prediction class subordinate sentence in report to security to be analyzed and scores；
Computing module is calculated to be analyzed for being ground the scoring of each prediction class subordinate sentence in report based on security to be analyzed
Security grind the whole emotion score of report；
Sentiment analysis result obtains module, and the whole emotion score and preset fraction threshold value of report are ground for being analysed to security
It is compared, and the sentiment analysis result that security to be analyzed grind report is obtained according to comparison result.
Further, the sentiment dictionary is established, is included the following steps:
Step 110, report is ground to the security of preset record and carries out subordinate sentence processing, obtained security and grind each first subordinate sentence in reporting；
Step 120, word segmentation processing is carried out to each first subordinate sentence, obtains security and grinds each first participle in reporting；
Step 130, it extracts from all first participles for expressing the participle of positive emotion, for expressing negative emotion
Participle, the participle for expressing negative emotion and the participle for expressing prediction；
Step 140, according to extract result generate respectively positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and
Predict class dictionary；
Step 150, institute is established based on positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction class dictionary
State sentiment dictionary.
Further, the prediction class subordinate sentence determining module, is specifically used for:
Each participle that security are ground in report is analysed to be matched with the participle in the sentiment dictionary in prediction class dictionary；
It can be with the participle phase in the sentiment dictionary in prediction class dictionary if security to be analyzed grind any participle in report
Subordinate sentence belonging to any participle is then determined as predicting class subordinate sentence by matching.
Further, the affective style determining module, is specifically used for:
Be analysed to security grind report in prediction class subordinate sentence in it is each segment respectively with positive emotion in the sentiment dictionary
Dictionary, negative emotion dictionary, negative emotion dictionary in participle matched；
It can be with positive emotion dictionary, negative emotion dictionary or negative feelings if security to be analyzed grind any participle in report
Participle in sense dictionary matches, then the affective style of any participle is determined as front participle, negative participle or negative
Participle.
Further, institute's scoring module, comprising:
Statistic unit grinds each front participle predicted in class subordinate sentence, negative participle in report for counting security to be analyzed
Initial number and negative participle quantity；
First judging unit, for judging whether the quantity of negative participle is odd number；
Second judgment unit, for when the judging result of the first judging unit, which is, is, then judging odd number negative participle
Above or below it is adjacent be front participle or negative participle；
Segment number adjustment unit, for when odd number negate participle above or below it is adjacent be front participle when, then
The initial number of front participle is subtracted, the initial number negatively segmented is added one together, to be segmented as front adjusted
Quantity and negatively segment quantity；When odd number negate participle above or below it is adjacent be negatively to segment when, then will negatively segment
Initial number subtract the initial number of front participle added one together, using as negative participle quantity adjusted and front
Segment quantity；
Score unit, grinds front participle quantity and negative participle in each prediction class subordinate sentence in report for being analysed to security
The difference of quantity divided by front participle quantity and negative participle quantity and, grind each prediction class in report to obtain security to be analyzed and divide
The scoring of sentence.
Further, it includes just dividing and bearing point that security to be analyzed, which grind the scoring of each prediction class subordinate sentence in report, is respectively corresponded
Front prediction class subordinate sentence and negative prediction class subordinate sentence.
Based on this, the computing module is specifically used for:
It calculates separately security to be analyzed and grinds all positive average marks for predicting class subordinate sentences in reporting and all negative prediction classes
The average mark of subordinate sentence；
It is analysed to the average mark and all negative prediction class subordinate sentences of all front prediction class subordinate sentences that security are ground in report
Two times of progress read group totals of average mark, to obtain the whole emotion score that security to be analyzed grind report.
Further, the preset fraction threshold value include the first preset fraction threshold value and the second preset fraction threshold value, first
Preset fraction threshold value is less than the second preset fraction threshold value.
Based on this, sentiment analysis result obtains module, is specifically used for:
By whole emotion score lower than the first preset fraction threshold value, between the first preset fraction threshold value and the second preset fraction
Investment securities to be analyzed between threshold value, higher than the second preset fraction threshold value grind the sentiment analysis of report as a result, being identified as seeing
Sky, level watching, see it is more.
To achieve the goals above, the present invention also provides a kind of computer equipments, including memory, processor and storage
On a memory and the computer program that can run on a processor, the processor realize the above method when executing described program
The step of.
To achieve the goals above, the present invention also provides a kind of computer readable storage medium, it is stored thereon with computer
Program, when described program is executed by processor the step of the realization above method.
Security provided by the invention using big data grind report sentiment analysis method, apparatus, computer equipment and readable deposit
Storage media can grind the participle in report to a large amount of security in advance and extract to generate the emotion word for including required affective style dictionary
Allusion quotation is based on this, can first be analysed to security and grind report progress subordinate sentence and word segmentation processing, then each participle and feelings by that will obtain
The mode of sense dictionary matching first determines that prediction class subordinate sentence determines the affective style respectively segmented in prediction class subordinate sentence again, next
It to each prediction class subordinate sentence is scored based on the affective style respectively segmented in prediction class subordinate sentence and using preset code of points, and can
The whole emotion score that security to be analyzed grind report is calculated based on appraisal result, the entirety of report is finally ground according to security to be analyzed
Emotion score and the comparison result of preset score threshold obtain the sentiment analysis result that security to be analyzed grind report.By this programme,
The sentiment analysis that security to be analyzed are ground with report can be realized based on the mode of big data analysis and intelligent scoring, can not only be saved significantly
Human-saving, and analysis efficiency can be improved and analyze the accuracy rate of result.
Detailed description of the invention
Fig. 1 is the optional stream of one kind that the security according to an embodiment of the present invention using big data grind report sentiment analysis method
Journey schematic diagram；
Fig. 2 is a kind of optional journey that the security according to an embodiment of the present invention using big data grind report sentiment analysis device
Sequence module diagram；
Fig. 3 is that the security according to an embodiment of the present invention using big data grind the another kind of report sentiment analysis device optionally
Program module schematic diagram；
Fig. 4 is a kind of optional hardware structure schematic diagram of computer equipment according to an embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.Based on the embodiments of the present invention, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall within the protection scope of the present invention.
Embodiment one
Report sentiment analysis method is ground to the security provided by the invention using big data with reference to the accompanying drawing to be illustrated.
Fig. 1 is a kind of optional flow diagram that the present invention grinds report sentiment analysis method using the security of big data, such as
Shown in Fig. 1, this method be may comprise steps of:
S01, the security to be analyzed for receiving input grind report.
Sentiment analysis is carried out, when receiving one
After piece security to be analyzed grind report, the format that first can grind report to this security to be analyzed judges.Specifically, can determine whether this wait divide
Analysis security grind whether the format of report is text type, such as " .txt " format, " .doc " format etc..If judging result be it is no,
That is the format that the security to be analyzed grind report is not text type, then the format that the security to be analyzed can be ground report is converted to text
This type, for example it is portable document format (PDF format) that security, which grind the format of report, and existing " PDFParser " etc. may be used
The security that the security of PDF format grind report conversion txt format are ground report by tool.With this, the security to be analyzed that can uniformly receive are ground
The format of report is read out with the more convenient and more efficient content for grinding report to security to be analyzed.
S02 grinds report to security to be analyzed and carries out subordinate sentence processing, obtains security to be analyzed and grinds each subordinate sentence in reporting.
In the present embodiment, report can be ground to security to be analyzed according to the symbol of preset type and carries out subordinate sentence processing, such as can
According to comma, ", fullstop ".", point number " ", dash "-", bracket " [], [] ", branch "；" etc. symbols, to be analyzed
Security grind report and carry out subordinate sentence processing, grind each subordinate sentence in reporting to obtain security to be analyzed.
S03 grinds each subordinate sentence in report to security to be analyzed and carries out word segmentation processing, obtains security to be analyzed and grinds each point in reporting
Word.
It in specific implementation, can be further using in the prior art after obtaining each subordinate sentence that security to be analyzed are ground in report
Word segmentation module, for example word segmentation processing is carried out to above-mentioned each subordinate sentence using jieba word segmentation module, grinds report to obtain security to be analyzed
In each participle.
S04 grinds each participle in report based on security to be analyzed, determines security to be analyzed using the sentiment dictionary pre-established
Grind the prediction class subordinate sentence in report.
Firstly, can first be illustrated to the establishment process of sentiment dictionary, which may include following steps:
Step 110, report is ground to the security of preset record and carries out subordinate sentence processing, obtained security and grind each subordinate sentence in reporting (at this
In embodiment, the first subordinate sentence can be described as).
Preset record (such as 300 etc.) is chosen in advance being related to the security of different industries and grind report, then each piece security are ground
Report carries out subordinate sentence processing, grinds each first subordinate sentence in reporting to obtain all security.
Step 120, word segmentation processing is carried out to each first subordinate sentence, obtains security and grinds each first participle in reporting.
After obtaining each first subordinate sentence, using existing word segmentation module, such as using jieba word segmentation module to each first point
Sentence carries out word segmentation processing, grinds each participle (in the present embodiment, can be described as the first participle) in reporting to obtain all security.
Step 130, it extracts from all first participles for expressing the participle of positive emotion, for expressing negative emotion
Participle, the participle for expressing negative emotion and the participle for expressing prediction.
After obtaining each first participle that all security are ground in report, we are combined using artificial extract with machine extraction
Mode, (mainly include to word more in terms of market and industry from the participle extracted in each first participle for expressing positive emotion
Language, such as " leading big city ", " Underpricing " etc.), the participle for expressing negative emotion (mainly include being seen to market and industry
Empty word, such as " economic recession ", " barely satisfactory " etc.), for express negate that the participle of emotion (refers mainly to that sentence can be contained
Justice plays the word of reversal effect, such as " extremely difficult ", " groundless " etc.), for express prediction participle (refer mainly to author's
Anticipation such as suggests at the relevant word, is mainly used for subsequent location prediction class subordinate sentence, such as " it is recommended that ", " it is expected that ", " prediction " etc.),
In addition, in practical applications, can also further extract participle relevant to industry and (be mainly used for positioning industry, such as " coloured
Metal ", " chemical industry " etc.).
Certainly, after the completion of extraction, the step of desk checking can also be added, is verified to result is extracted, to obtain
More accurate extraction to be subsequent as a result, to establish each related dictionary and do sufficient data preparation.
Step 140, according to extract result generate respectively positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and
Predict class dictionary.
That is, positive emotion dictionary can be generated according to the participle for expressing positive emotion of extraction, according to extraction
The participle for expressing negative emotion generate negative emotion dictionary, according to extraction for express the participle generation of negative emotion
Negate emotion dictionary, prediction class libraries is generated according to the participle for expressing prediction of extraction, according to relevant point of the industry of extraction
Word generates the similar dictionary of industry.
Step 150, feelings are established based on positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction class dictionary
Feel dictionary.
By the positive emotion dictionary of generation, negative emotion dictionary, negative emotion dictionary, prediction class dictionary, the similar word of industry
Library etc. is integrated to establish sentiment dictionary, can be according to sentiment dictionary so as to subsequent when grinding report progress sentiment analysis to security
In each dictionary fast and accurately determine participle type, subordinate sentence type etc., sentence, article etc. are made subject to more to realize
True analysis (especially sentiment analysis).
In the present embodiment, after S03 obtains each participle that security to be analyzed are ground in report, security can be analysed to and ground in report
It is each participle matched with the participle in the prediction class dictionary in sentiment dictionary, if security to be analyzed grind report in any participle
It can match with the participle in sentiment dictionary in prediction class dictionary, then any participle can be determined as predicting class participle, into
And subordinate sentence belonging to any participle can be determined as predicting class subordinate sentence.
Due to being ground in report in security, the ground of prediction quasi-sentence (also being understood as the prediction class subordinate sentence in this programme) appearance
Side usually corresponds to author to be used to express to the studying and judging of future market market development (for example to may include being expected to rise, seeing that the emotions such as sky become
Gesture), therefore, first determine prediction class subordinate sentence, then sentiment analysis is carried out to the participle in prediction class subordinate sentence, it can be accurate and effective
Analyze the emotion trend of author.
S05 determines that security to be analyzed grind the affective style respectively segmented in the prediction class subordinate sentence in report according to sentiment dictionary.
In specific implementation, after each prediction class subordinate sentence has been determined, it can will predict that each participle for including in class subordinate sentence is distinguished
It is matched with the participle in positive emotion dictionary in sentiment dictionary, negative emotion dictionary, negative emotion dictionary, if any participle
It can match with the participle in positive emotion dictionary, or can match with the participle in negative emotion dictionary, Huo Zheneng
The participle felt in dictionary of pledging love whether enough matches, then the affective style of any participle can be identified as front participle,
Negative participle, negative participle.
S06 is ground the affective style respectively segmented in the prediction class subordinate sentence in report based on security to be analyzed, is advised using preset scoring
Then each prediction class subordinate sentence in report is ground to security to be analyzed to score.
After the affective style respectively segmented in prediction class subordinate sentence has been determined, then it can be determined based on the affective style of each participle
The emotion trend of affiliated each subordinate sentence can become to the emotion of each prediction class subordinate sentence in a manner of intelligent scoring in the present embodiment
Gesture is evaluated, and gained can score and be applied to subsequent step so that the whole emotion score that security grind report is calculated.
In specific implementation, the front that security to be analyzed are ground in report in each prediction class subordinate sentence can be first counted to segment, is negative
The initial number (since the quantity may be adjusted, referred to herein as initial number) of participle, and the number of negative participle
Amount.
By taking a prediction class subordinate sentence as an example, such as the initial number that can in the front participle for counting the subordinate sentence, negatively segment
After amount (for example being respectively 3 and 2), judge whether the quantity of negative participle is odd number.
If the quantity of negative participle is odd number (for example, the quantity of negative participle is 1, being " being difficult "), then can be into one
Step judge the odd number negative participle front is adjacent or rear adjacent is front participle or negatively segments.
If front is adjacent or rear adjacent is front participle (such as " leading big city ") for odd number negative participle,
Then by negative participle, with front participle, combine expression is the negative meaning, at this time, it may be necessary to by the front segment at the beginning of
Beginning quantity subtracts one, and (i.e. 3-1=2) and the initial number that negatively segments this add one (i.e. 2+1=3), using as adjusted
Front segments quantity and negatively segments quantity.
If front is adjacent or rear adjacent is negative participle (such as " drop ") for odd number negative participle, should
With the negative participle, combine expression is the positive meaning to negative participle, at this time, it may be necessary to the initial number that this is negatively segmented
Subtracting one (i.e. 2-1=1) and adds one (i.e. 3+1=4) for the initial number that the front segments, as negative point adjusted
Word quantity and front participle quantity.
By above-mentioned odd number negate participle rear adjacent be negative participle in case where, determined it is adjusted just
It, can be by the difference of front participle quantity in the prediction class subordinate sentence and negative participle quantity after face segments quantity and negative participle quantity
(i.e. 4-1=3), divided by front participle quantity and negative participle quantity and (i.e. 4+1=5), to obtain to the prediction class subordinate sentence
Scoring (i.e. 3/5=0.6).
It is usually ground in report in security, predicting that the emotion trend of class subordinate sentence and security grind the whole emotion trend of report is breath manner of breathing
It closes, therefore, the whole emotion trend that security grind report can be obtained in the next steps based on the scoring of each prediction class subordinate sentence.
S07 is ground the scoring of each prediction class subordinate sentence in report based on security to be analyzed, security to be analyzed is calculated and grind report
Whole emotion score.
In the present embodiment, it includes just dividing and bear point that security to be analyzed, which grind the scoring of each prediction class subordinate sentence in report, is distinguished
Corresponding front prediction class subordinate sentence and negative prediction class subordinate sentence.
S06 be calculated security to be analyzed grind report in each prediction class subordinate sentence scoring after, which can be carried out
Assign power accumulation calculating.
Specifically, the average mark for all front prediction class subordinate sentences that security to be analyzed are ground in report, Yi Jisuo can be calculated separately
There is the average mark of negative prediction class subordinate sentence, since in a practical situation, most security grind report and are all biased into positive viewpoint,
Therefore, in the present solution, we assign higher weight to negative prediction class subordinate sentence, for example it may be configured as front prediction class subordinate sentence
Two times.
It, can be by the average mark of the average marks of all front prediction class subordinate sentences and all negative prediction class subordinate sentences when specific calculating
Two times of progress read group totals, to obtain the whole emotion score that the security to be analyzed grind report, specifically can by following formula into
Row calculates:
The whole emotion score=all fronts prediction class subordinate sentence average mark+(average mark of all negative prediction class subordinate sentences
×2)。
S08, is analysed to security and grinds the whole emotion score of report to be compared with preset fraction threshold value, and ties according to comparing
Fruit obtains the sentiment analysis result that security to be analyzed grind report.
Theoretically, which may be configured as 0, but in view of in a practical situation, most of security grind report all
It is biased into positive viewpoint, therefore, in the present embodiment, which may be configured as including the first preset fraction threshold value
With the second preset fraction threshold value, and the first preset fraction threshold value is less than the second preset fraction threshold value, for example can be respectively set to 0 He
0.2。
In specific implementation, can by whole emotion score lower than the first preset fraction threshold value (namely be lower than 0) to point
The sentiment analysis result that analysis security grind report is determined as seeing sky；Whole emotion score is pre- with second between the first preset fraction threshold value
If the sentiment analysis result that the security to be analyzed between score threshold (namely between 0 to 0.2) grind report is determined as level watching；It will
The sentiment analysis of report is ground in the investment securities to be analyzed that whole emotion score is higher than the second preset fraction threshold value (being namely higher than 0.2)
As a result it is determined as seeing more (being also referred to as expected to rise).
With this, can by by objective appraisal result compared with desired indicator pair, to obtain the feelings that security to be analyzed grind report
Sense analysis is as a result, to can guarantee the objectivity and accuracy of the sentiment analysis result.
According to each embodiment of the present embodiment, the participle in report can be ground to a large amount of security in advance and extracted to generate
Sentiment dictionary including required affective style dictionary is based on this, can first be analysed to security and grind at report progress subordinate sentence and participle
Then reason first determines that prediction class subordinate sentence determines prediction class again with the matched mode of sentiment dictionary by each participle that will be obtained
Next the affective style respectively segmented in subordinate sentence is advised based on the affective style for predicting respectively to segment in class subordinate sentence and using preset scoring
It then scores each prediction class subordinate sentence, and the whole emotion point that security to be analyzed grind report can be calculated based on appraisal result
Number finally grinds the whole emotion score of report and the comparison result acquisition security to be analyzed of preset score threshold according to security to be analyzed
Grind the sentiment analysis result of report.By this programme, can be realized based on the mode of big data analysis and intelligent scoring to be analyzed
Security grind the sentiment analysis of report, can not only greatly save manpower, and analysis efficiency can be improved and analyze the accuracy rate of result.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.
Embodiment two
The security using big data provided in one based on the above embodiment grind report sentiment analysis method, mention in the present embodiment
Report sentiment analysis device is ground for a kind of security using big data, specifically, Fig. 2 to 3 shows the security for utilizing big data
The optional structural block diagram of report sentiment analysis device is ground, which grinds report sentiment analysis device and be divided into one
A or multiple program modules, one or more program module are stored in storage medium, and by one or more processors
It is performed, to complete the present invention.The so-called program module of the present invention is the series of computation machine journey for referring to complete specific function
Sequence instruction segment is more suitable for describing grinding report sentiment analysis device holding in storage medium using the security of big data than program itself
Row process, the function of each program module of the present embodiment will specifically be introduced by being described below.
As shown in Fig. 2, the security using big data grind report sentiment analysis device 20 can include:
Receiving module 21, the security to be analyzed that can be used for receiving input grind report；
Subordinate sentence module 22 can be used for grinding security to be analyzed report and carry out subordinate sentence processing, obtains security to be analyzed and grinds in report
Each subordinate sentence；
Word segmentation module 23 can be used for grinding security to be analyzed each subordinate sentence in report and carry out word segmentation processing, obtains card to be analyzed
Certificate grinds each participle in report；
It predicts class subordinate sentence determining module 24, can be used for grinding each participle in report based on security to be analyzed, using pre-establishing
Sentiment dictionary determine security to be analyzed grind report in prediction class subordinate sentence；
Affective style determining module 25 can be used for determining that security to be analyzed grind the prediction class in report according to the sentiment dictionary
The affective style respectively segmented in subordinate sentence；
Grading module 26 can be used for grinding the affective style respectively segmented in the prediction class subordinate sentence in report based on security to be analyzed,
Each prediction class subordinate sentence in report is ground to security to be analyzed using preset code of points to score；
Computing module 27, can be used for grinding based on security to be analyzed report in each prediction class subordinate sentence scoring, be calculated to
Analysis security grind the whole emotion score of report；
Sentiment analysis result obtains module 28, can be used for being analysed to whole emotion score and preset fraction that security grind report
Threshold value is compared, and obtains the sentiment analysis result that security to be analyzed grind report according to comparison result.
In the present embodiment, sentiment dictionary is established, it may include following steps:
Step 110, report is ground to the security of preset record and carries out subordinate sentence processing, obtained security and grind each first subordinate sentence in reporting；
Step 120, word segmentation processing is carried out to each first subordinate sentence, obtains security and grinds each first participle in reporting；
Step 130, it extracts from all first participles for expressing the participle of positive emotion, for expressing negative emotion
Participle, the participle for expressing negative emotion and the participle for expressing prediction；
Step 140, according to extract result generate respectively positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and
Predict class dictionary；
Step 150, feelings are established based on positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction class dictionary
Feel dictionary.
Further, it predicts class subordinate sentence determining module 24, can be specifically used for:
Each participle that security are ground in report is analysed to be matched with the participle in the sentiment dictionary in prediction class dictionary；
It can be with the participle phase in the sentiment dictionary in prediction class dictionary if security to be analyzed grind any participle in report
Subordinate sentence belonging to any participle is then determined as predicting class subordinate sentence by matching.
Further, affective style determining module 25 can be specifically used for:
Be analysed to security grind report in prediction class subordinate sentence in it is each segment respectively with positive emotion in the sentiment dictionary
Dictionary, negative emotion dictionary, negative emotion dictionary in participle matched；
It can be with positive emotion dictionary, negative emotion dictionary or negative feelings if security to be analyzed grind any participle in report
Participle in sense dictionary matches, then the affective style of any participle is determined as front participle, negative participle or negative
Participle.
In specific implementation, shown in referring to Fig. 3, grading module 26 be may particularly include:
Statistic unit 261 can be used for counting the front that security to be analyzed are ground in report in each prediction class subordinate sentence and segment, be negative
The initial number of participle and the quantity of negative participle；
First judging unit 262 can be used for judging whether the quantity of negative participle is odd number；
Second judgment unit 263 can be used for then judging odd number when the judging result of the first judging unit 262, which is, is
Negative participle above or below it is adjacent be front participle or negative participle；
Segment number adjustment unit 264, can be used for when odd number negative participle above or below it is adjacent be positive participle
When, then the initial number of front participle is subtracted the initial number negatively segmented is added one together, using as it is adjusted just
Face segments quantity and negatively segments quantity；When odd number negate participle above or below it is adjacent be negatively to segment when, then will bear
The initial number of face participle, which subtracts, adds one for the initial number of front participle together, using as negative participle quantity adjusted
And front participle quantity；
Scoring unit 265 can be used for being analysed to security and grind front participle quantity in each prediction class subordinate sentence in reporting and bear
Face segment the difference of quantity divided by front participle quantity and negative participle quantity and, with obtain security to be analyzed grind report in it is each pre-
Survey the scoring of class subordinate sentence.
In specific implementation, it includes just dividing and bear point that security to be analyzed, which grind the scoring of each prediction class subordinate sentence in report, is distinguished
Corresponding front prediction class subordinate sentence and negative prediction class subordinate sentence.
Based on this, computing module 27 can be specifically used for:
It calculates separately security to be analyzed and grinds all positive average marks for predicting class subordinate sentences in reporting and all negative prediction classes
The average mark of subordinate sentence；
It is analysed to the average mark and all negative prediction class subordinate sentences of all front prediction class subordinate sentences that security are ground in report
Two times of progress read group totals of average mark, to obtain the whole emotion score that security to be analyzed grind report.
Further, preset fraction threshold value may include the first preset fraction threshold value and the second preset fraction threshold value, and first is pre-
If score threshold is less than the second preset fraction threshold value.
Based on this, sentiment analysis result obtains module 28, can be specifically used for:
By whole emotion score lower than the first preset fraction threshold value, between the first preset fraction threshold value and the second preset fraction
Investment securities to be analyzed between threshold value, higher than the second preset fraction threshold value grind the sentiment analysis of report as a result, being identified as seeing
Sky, level watching, see it is more.
About the device in above-described embodiment, wherein each unit, module execute the concrete mode of operation related
It is described in detail in the embodiment of this method, no detailed explanation will be given here.
Each embodiment through this embodiment can grind the participle in report to a large amount of security in advance and extract to generate
Sentiment dictionary including required affective style dictionary is based on this, can first be analysed to security and grind at report progress subordinate sentence and participle
Then reason first determines that prediction class subordinate sentence determines prediction class again with the matched mode of sentiment dictionary by each participle that will be obtained
Next the affective style respectively segmented in subordinate sentence is advised based on the affective style for predicting respectively to segment in class subordinate sentence and using preset scoring
It then scores each prediction class subordinate sentence, and the whole emotion point that security to be analyzed grind report can be calculated based on appraisal result
Number finally grinds the whole emotion score of report and the comparison result acquisition security to be analyzed of preset score threshold according to security to be analyzed
Grind the sentiment analysis result of report.By this programme, can be realized based on the mode of big data analysis and intelligent scoring to be analyzed
Security grind the sentiment analysis of report, can not only greatly save manpower, and analysis efficiency can be improved and analyze the accuracy rate of result.
Embodiment three
The present embodiment also provides a kind of computer equipment, can such as execute the smart phone, tablet computer, notebook of program
Computer, desktop computer, rack-mount server, blade server, tower server or Cabinet-type server are (including independent
Server cluster composed by server or multiple servers) etc..As shown in figure 4, the computer equipment 40 of the present embodiment to
It is few to include but is not limited to: memory 41, the processor 42 of connection can be in communication with each other by system bus, as shown in Figure 4.It needs to refer to
Out, Fig. 4 illustrates only the computer equipment 40 with component 41-42, it should be understood that being not required for implementing all
The component shown, the implementation that can be substituted is more or less component.
In the present embodiment, memory 41 (i.e. readable storage medium storing program for executing) includes flash memory, hard disk, multimedia card, card-type memory
(for example, SD or DX memory etc.), random access storage device (RAM), static random-access memory (SRAM), read-only memory
(ROM), electrically erasable programmable read-only memory (EEPROM), programmable read only memory (PROM), magnetic storage, magnetic
Disk, CD etc..In some embodiments, memory 41 can be the internal storage unit of computer equipment 40, such as the calculating
The hard disk or memory of machine equipment 40.In further embodiments, memory 41 is also possible to the external storage of computer equipment 40
The plug-in type hard disk being equipped in equipment, such as the computer equipment 40, intelligent memory card (Smart Media Card, SMC), peace
Digital (Secure Digital, SD) card, flash card (Flash Card) etc..Certainly, memory 41 can also both include meter
The internal storage unit for calculating machine equipment 40 also includes its External memory equipment.In the present embodiment, memory 41 is commonly used in storage
It is installed on the operating system and types of applications software of computer equipment 40, such as the security using big data of embodiment two grind report
The program code etc. of sentiment analysis device.In addition, memory 41 can be also used for temporarily storing and export or will be defeated
Various types of data out.
Processor 42 can be in some embodiments central processing unit (Central Processing Unit, CPU),
Controller, microcontroller, microprocessor or other data processing chips.The processor 42 is commonly used in control computer equipment
40 overall operation.In the present embodiment, program code or processing data of the processor 42 for being stored in run memory 41,
Such as report sentiment analysis device etc. is ground using the security of big data.
Example IV
The present embodiment also provides a kind of computer readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory
(for example, SD or DX memory etc.), random access storage device (RAM), static random-access memory (SRAM), read-only memory
(ROM), electrically erasable programmable read-only memory (EEPROM), programmable read only memory (PROM), magnetic storage, magnetic
Disk, CD, server, App are stored thereon with computer program, phase are realized when program is executed by processor using store etc.
Answer function.The computer readable storage medium of the present embodiment is used to grind report sentiment analysis device using the security of big data, is located
Reason device realizes that the security using big data of embodiment one grind report sentiment analysis method when executing.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of security using big data grind report sentiment analysis method, which comprises the steps of:
S01, the security to be analyzed for receiving input grind report；
S02 grinds report to security to be analyzed and carries out subordinate sentence processing, obtains security to be analyzed and grinds each subordinate sentence in reporting；
S03 grinds each subordinate sentence in report to security to be analyzed and carries out word segmentation processing, obtains security to be analyzed and grinds each participle in reporting；
S04 grinds each participle in report based on security to be analyzed, determines that security to be analyzed grind report using the sentiment dictionary pre-established
In prediction class subordinate sentence；
S05 determines that security to be analyzed grind the affective style respectively segmented in the prediction class subordinate sentence in report according to the sentiment dictionary；
S06 is ground the affective style respectively segmented in the prediction class subordinate sentence in report based on security to be analyzed, utilizes preset code of points pair
Each prediction class subordinate sentence that security to be analyzed are ground in report scores；
S07 is ground the scoring of each prediction class subordinate sentence in report based on security to be analyzed, the entirety that security to be analyzed grind report is calculated
Emotion score；
S08, is analysed to security and grinds the whole emotion score of report to be compared with preset fraction threshold value, and is obtained according to comparison result
Security to be analyzed are taken to grind the sentiment analysis result of report.
2. the security according to claim 1 using big data grind report sentiment analysis method, which is characterized in that described in foundation
Sentiment dictionary includes the following steps:
Step 110, report is ground to the security of preset record and carries out subordinate sentence processing, obtained security and grind each first subordinate sentence in reporting；
Step 120, word segmentation processing is carried out to each first subordinate sentence, obtains security and grinds each first participle in reporting；
Step 130, the participle for expressing positive emotion, point for expressing negative emotion are extracted from all first participles
Word, the participle for expressing negative emotion and the participle for expressing prediction；
Step 140, positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction are generated according to extraction result respectively
Class dictionary；
Step 150, the feelings are established based on positive emotion dictionary, negative emotion dictionary, negative emotion dictionary and prediction class dictionary
Feel dictionary.
3. the security according to claim 2 using big data grind report sentiment analysis method, which is characterized in that S04 is based on
Security to be analyzed grind each participle in report, determine that security to be analyzed grind the prediction class point in report using the sentiment dictionary pre-established
Sentence, comprising:
Each participle that security are ground in report is analysed to be matched with the participle in the sentiment dictionary in prediction class dictionary；
If security to be analyzed, which grind any participle in report, to match with the participle in the sentiment dictionary in prediction class dictionary,
Then subordinate sentence belonging to any participle is determined as to predict class subordinate sentence.
4. the security according to claim 2 using big data grind report sentiment analysis method, which is characterized in that S05 according to
The sentiment dictionary determines that security to be analyzed grind the affective style respectively segmented in the prediction class subordinate sentence in report, comprising:
Be analysed to that security grind in prediction class subordinate sentence in report it is each segment respectively with positive emotion dictionary in the sentiment dictionary,
Participle in negative emotion dictionary, negative emotion dictionary is matched；
It can be with positive emotion dictionary, negative emotion dictionary or negative emotion word if security to be analyzed grind any participle in report
Participle in library matches, then the affective style of any participle is determined as front participle, negative participle or negative and segmented.
5. the sentiment analysis that security according to claim 4 grind report grinds report sentiment analysis method using the security of big data,
It is characterized in that, S06 grinds the affective style respectively segmented in the prediction class subordinate sentence in report based on security to be analyzed, preset scoring is utilized
Rule grinds each prediction class subordinate sentence in report to security to be analyzed and scores, comprising:
It counts security to be analyzed and grinds initial number and negative point that the front in report in each prediction class subordinate sentence segments, negatively segments
The quantity of word；
Whether the quantity for judging negative participle is odd number；
If so, judge odd number negative participle above or below it is adjacent be front participle or negative participle；
If odd number negative participle above or below it is adjacent be positive participle, by front participle initial number subtract together will
The initial number negatively segmented adds one, to segment quantity as front adjusted and negatively segment quantity；If odd number
Negative participle above or below it is adjacent be negatively to segment, then by the initial number negatively segmented subtract together by front participle just
Beginning quantity adds one, to segment quantity as negative participle quantity adjusted and front；
Security are analysed to grind front participle quantity in each prediction class subordinate sentence in reporting and negatively segment the difference of quantity divided by front
Segment quantity and negative participle quantity and, to obtain the scoring that security to be analyzed grind each prediction class subordinate sentence in reporting.
6. the security according to claim 1 using big data grind report sentiment analysis method, which is characterized in that card to be analyzed
The scoring that certificate grinds each prediction class subordinate sentence in report includes just dividing and bear point, respectively corresponds front prediction class subordinate sentence and negatively predicts class
Subordinate sentence；
S07 grinds the scoring of each prediction class subordinate sentence in report based on security to be analyzed, and the entirety that security to be analyzed grind report is calculated
Emotion score, comprising:
Calculate separately the average mark and all negative prediction class subordinate sentences of all front prediction class subordinate sentences that security to be analyzed are ground in report
Average mark；
It is analysed to all fronts that security are ground in report and predicts the average mark of class subordinate sentences and being averaged for all negative prediction class subordinate sentences
The two times of progress read group totals divided, to obtain the whole emotion score that security to be analyzed grind report.
7. the security according to claim 1 using big data grind report sentiment analysis method, which is characterized in that described default
Score threshold includes the first preset fraction threshold value and the second preset fraction threshold value, and the first preset fraction threshold value is less than second default point
Number threshold value；
S08 is analysed to security and grinds the whole emotion score of report to be compared with preset fraction threshold value, and is obtained according to comparison result
Security to be analyzed are taken to grind the sentiment analysis result of report, comprising:
By whole emotion score lower than the first preset fraction threshold value, between the first preset fraction threshold value and the second preset fraction threshold value
Between, the investment securities to be analyzed higher than the second preset fraction threshold value grind the sentiment analysis of report as a result, be identified as seeing it is empty, see
Put down, see it is more.
8. a kind of security using big data grind report sentiment analysis device characterized by comprising
Receiving module, security to be analyzed for receiving input grind report；
Subordinate sentence module carries out subordinate sentence processing for grinding report to security to be analyzed, obtains security to be analyzed and grinds each subordinate sentence in reporting；
Word segmentation module, each subordinate sentence for being ground in report to security to be analyzed carry out word segmentation processing, obtain security to be analyzed and grind in report
Each participle；
It predicts class subordinate sentence determining module, for grinding each participle in report based on security to be analyzed, utilizes the emotion word pre-established
Allusion quotation determines that security to be analyzed grind the prediction class subordinate sentence in report；
Affective style determining module, it is each in the prediction class subordinate sentence in report for determining that security to be analyzed are ground according to the sentiment dictionary
The affective style of participle；
Grading module, for grinding the affective style that respectively segments in each prediction class subordinate sentence in report based on security to be analyzed, using pre-
Set code of points to security to be analyzed grind report in each prediction class subordinate sentence score；
Security to be analyzed are calculated for grinding the scoring of each prediction class subordinate sentence in report based on security to be analyzed in computing module
Grind the whole emotion score of report；
Sentiment analysis result obtains module, and for being analysed to, security grind the whole emotion score of report and preset fraction threshold value carries out
It compares, and the sentiment analysis result that security to be analyzed grind report is obtained according to comparison result.
9. a kind of computer equipment, the computer equipment include memory, processor and storage on a memory and can be
The computer program run on processor, which is characterized in that the processor realizes claim 1 to 7 when executing described program
The step of any one the method.
10. a kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that: described program is processed
The step of any one of claim 1 to 7 the method is realized when device executes.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811228240.1A CN109460550A (en) | 2018-10-22 | 2018-10-22 | Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data |
| PCT/CN2019/070291 WO2020082612A1 (en) | 2018-10-22 | 2019-01-03 | Method and apparatus for sentiment analysis on security research report using big data, and computer device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811228240.1A CN109460550A (en) | 2018-10-22 | 2018-10-22 | Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109460550A true CN109460550A (en) | 2019-03-12 |
Family
ID=65608037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811228240.1A Pending CN109460550A (en) | 2018-10-22 | 2018-10-22 | Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109460550A (en) |
| WO (1) | WO2020082612A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110295A (en) * | 2019-04-04 | 2019-08-09 | 平安科技（深圳）有限公司 | Large sample grinds report information extracting method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060242040A1 (en) * | 2005-04-20 | 2006-10-26 | Aim Holdings Llc | Method and system for conducting sentiment analysis for securities research |
| CN107767273A (en) * | 2017-09-05 | 2018-03-06 | 平安科技（深圳）有限公司 | Asset Allocation method, electronic installation and medium based on social data |
| CN107885883A (en) * | 2017-12-01 | 2018-04-06 | 北京国信宏数科技有限公司 | A kind of macroeconomy field sentiment analysis method and system based on Social Media |
| CN108062300A (en) * | 2016-11-08 | 2018-05-22 | 中移(苏州)软件技术有限公司 | A kind of method and device that Sentiment orientation analysis is carried out based on Chinese text |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10515153B2 (en) * | 2013-05-16 | 2019-12-24 | Educational Testing Service | Systems and methods for automatically assessing constructed recommendations based on sentiment and specificity measures |
| CN103970864B (en) * | 2014-05-08 | 2017-09-22 | 清华大学 | Mood classification and mood component analyzing method and system based on microblogging text |
| CN108090040B (en) * | 2016-11-23 | 2021-08-17 | 北京国双科技有限公司 | Text information classification method and system |
| CN108629693A (en) * | 2018-05-08 | 2018-10-09 | 平安科技（深圳）有限公司 | Automatically generate method, apparatus, computer equipment and the storage medium of suggestion for investment |
-
2018
- 2018-10-22 CN CN201811228240.1A patent/CN109460550A/en active Pending
-
2019
- 2019-01-03 WO PCT/CN2019/070291 patent/WO2020082612A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060242040A1 (en) * | 2005-04-20 | 2006-10-26 | Aim Holdings Llc | Method and system for conducting sentiment analysis for securities research |
| CN108062300A (en) * | 2016-11-08 | 2018-05-22 | 中移(苏州)软件技术有限公司 | A kind of method and device that Sentiment orientation analysis is carried out based on Chinese text |
| CN107767273A (en) * | 2017-09-05 | 2018-03-06 | 平安科技（深圳）有限公司 | Asset Allocation method, electronic installation and medium based on social data |
| CN107885883A (en) * | 2017-12-01 | 2018-04-06 | 北京国信宏数科技有限公司 | A kind of macroeconomy field sentiment analysis method and system based on Social Media |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110295A (en) * | 2019-04-04 | 2019-08-09 | 平安科技（深圳）有限公司 | Large sample grinds report information extracting method, device, equipment and storage medium |
| CN110110295B (en) * | 2019-04-04 | 2023-10-20 | 平安科技（深圳）有限公司 | Large sample research and report information extraction method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020082612A1 (en) | 2020-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163478B (en) | Risk examination method and device for contract clauses | |
| CN109359175B (en) | Electronic device, litigation data processing method, and storage medium | |
| CN109815487B (en) | Text quality inspection method, electronic device, computer equipment and storage medium | |
| CN109492222B (en) | Intention identification method and device based on concept tree and computer equipment | |
| CN108595583A (en) | Dynamic chart class page data crawling method, device, terminal and storage medium | |
| CN111179066B (en) | Batch processing method and device for business data, server and storage medium | |
| CN107807941A (en) | Information processing method and device | |
| CN109101489A (en) | A kind of text automatic abstracting method, device and a kind of electronic equipment | |
| CN110427453B (en) | Data similarity calculation method, device, computer equipment and storage medium | |
| CN110489424A (en) | A kind of method, apparatus, storage medium and the electronic equipment of tabular information extraction | |
| CN111046184A (en) | Text risk identification method, device, server and storage medium | |
| CN109284504A (en) | It grinds to call the score using the security of deep learning model and analyses method and device | |
| CN111274390A (en) | Emotional reason determining method and device based on dialogue data | |
| CN110019556B (en) | Topic news acquisition method, device and equipment thereof | |
| CN109388804A (en) | Report core views extracting method and device are ground using the security of deep learning model | |
| CN109460550A (en) | Report sentiment analysis method, apparatus and computer equipment are ground using the security of big data | |
| CN111598122B (en) | Data verification method and device, electronic equipment and storage medium | |
| CN113362162A (en) | Wind control identification method and device based on network behavior data, electronic equipment and medium | |
| CN110689257B (en) | Operator big data-based fast-moving consumer goods business supervision method and device | |
| US20140250356A1 (en) | Method, device, and computer storage media for adding hyperlink to text | |
| CN116795978A (en) | Complaint information processing method and device, electronic equipment and medium | |
| CN110069772A (en) | Predict device, method and the storage medium of the scoring of question and answer content | |
| CN109446318A (en) | A kind of method and relevant device of determining auto repair document subject matter | |
| CN115757075A (en) | Task abnormity detection method and device, computer equipment and storage medium | |
| CN113569929B (en) | Internet service providing method and device based on small sample expansion and electronic equipment |
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 |