CN108550050A - A kind of user's portrait method based on call center data - Google Patents
A kind of user's portrait method based on call center data Download PDFInfo
- Publication number
- CN108550050A CN108550050A CN201810256594.0A CN201810256594A CN108550050A CN 108550050 A CN108550050 A CN 108550050A CN 201810256594 A CN201810256594 A CN 201810256594A CN 108550050 A CN108550050 A CN 108550050A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- portrait
- information
- neural network
- 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
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
In order to solve electric business, the real demand of client cannot be found out well by marketing or overcoming industry in most cases, the disadvantage of customer relationship can not well be handled, the present invention proposes a kind of user's portrait method based on call center data, pass through this method, this method leads to call center's network communication data by extraction, user is obtained to draw a portrait in the self information of special scenes, and accumulative a large amount of more scene communication datas, then the feature degree of user is described, transaction scene carries out retouching art, and itself fancy grade is described and analyzes, then it predicts the portrait of user by analysis result and generates data to take problem, form voice and questionnaire, it is final to realize precision marketing, improve user experience.
Description
Technical field
The invention belongs to big data digging technology field, more specifically, being related to a kind of based on call center data
User's portrait method.
Background technology
Data mining, also known as Knowledge Discovery in Database (Knowledge Discovery from Datebase, letter
Claim KDD), it is one and extracts the complex process for excavating the knowledge such as unknown, valuable pattern or rule from mass data.
In electric business, marketing or customer service industry, it usually needs customer relationship is handled by voice communication, but due to existing
Electric business, marketing or client's industry it is more strange to the other end of personage of voice, so in most cases cannot be fine
The real demand for findding out client, can not well handle customer relationship, and telemarketing now or network marketing then need to lead to
Customer relationship can be handled well by crossing voice, solve customer issue, and cannot generate unnecessary controversial issue.
Invention content
In view of this, the present invention proposes a kind of user's portrait method based on call center data, in this way, the party
Method leads to call center's network communication data by extraction, obtains user and draws a portrait in the self information of special scenes, and is accumulative a large amount of
More scene communication datas, then the feature degree of user is described, scene of merchandising retouch art and itself fancy grade into
Row description and analysis, then predict the portrait of user by analysis result and generate data to take problem, form voice and questionnaire,
It is final to realize precision marketing, improve user experience.
A kind of user's portrait method based on call center data comprising following steps:
S1:By changing into word to recording, and semantic data is extracted by NLPIR Chinese word segmentation systems;
S2:The data acquired in step S1 are cleaned;
S3:Labeling processing is extracted to the step S2 data cleaned;
S4:Clustering is carried out to the user tag in step S3, forms user's portrait;
S5:Predict Test and the words art of corresponding user is determined according to the analysis result in step S4 to supplement incomplete use
Family information material;
S6:The user information data in extraction supplement step S5 is excavated by communication data.
Further, step S1 is comprised the following specific steps that:
S11 first by the time with smaller spaced discrete, and records all calls in the call observation phase;
S12 is acquired the sound recordings and fileinfo of call center, and extracts user information and semantic information;
S13 arranges user information and semantic information, distinguishes user information, contextual data information and user's feelings
Thread information.
Further, the voice of phone system acquisition includes:The recording of customer service class, core body class voice, electricity pin class voice, collection
Class voice, questionnaire etc..
Further, the user information and semantic information extracted include:State, call mesh and demand point, call when
Between, the duration of call, customer site, IO circuits, number field, number information, connect the frequency, operation system information, client identity
Deng, satisfaction, name grammatical category information, keyword message:It such as swears at people, complain, quality inspection result, voice, intonation.Wherein, state, logical
Words purpose and demand point belong to contextual data information, and operation system information, client identity, number field, number information etc. belong to
In user information, and voice, intonation etc. then belong to user emotion information.
Further, step S2 is comprised the following specific steps that:
S21 carries out classification and scene analysis to collected data in S1 first;
S22 then clears up redundancy;
S23, the regularity then occurred to data are analyzed, and abnormal data is reasonably cleared data or change;
Further, step S3 includes the following steps:Keyword is extracted, and generates label, it is related to user for summarizing
The data such as hobby, individual character, feature.
Further, using BP neural network is arrived in step S4, which is a kind of multilayer feedforward neural network,
It may be implemented from the arbitrary nonlinear mapping for being input to output, have the characteristics that good self-organizing, adaptive.BP neural network
It can learn and store a large amount of input-output mode map relationship, without disclosing the mathematics for describing this mapping relations in advance
Equation.The learning rules used are to use steepest descent method, and the weights and threshold value of network are constantly adjusted by backpropagation, are made
The error sum of squares of network is minimum, terminates study.
Wherein, using with the present invention during, most important formula is as follows in BP neural network, enables P={ P1, P2...,
PmIndicate user's portrait set, wherein PiIndicate i-th of user's portrait.Ci={ Pi1, Pi2..., PinIndicate cluster after i-th
A label, wherein Pij, indicate CiJ-th of element in classification.After cluster, the condition that respectively classifying meet is Uk I=1Ci
=PCm≠Cr, thenCm∩Cr=φ, Min Δ p ∈ Cm, Δ pj ∈ Cr, Δ Cm,
(sim (Pi, Pj)) > Max Δs pi, pj ∈ Cm,Based on cluster result, it can be found that numerous
The information implied between miscellaneous user's portrait extracts new user's portrait label.
Further, step 4 includes the following steps:
S41 carries out feature information extraction first, from state, call mesh and demand point, air time, the duration of call, visitor
Family place, IO circuits, number field, number information, the connection frequency, operation system information, client identity etc., satisfaction, noun
Category information, keyword message:It such as swears at people, complain, choosing 11 in quality inspection result, voice, intonation as feature tag, together
When choose and carried out questionnaire survey, and marked the user of identity as training object, by 11 feature tags and user
Input information of the object as BP neural network;
S42, by the input information of BP neural network, including 11 feature tags and user object as training data and
11 feature tags and user object are input to BP neural network, for training by inspection data when as training data
Neural network takes learning rate η=0.3, error criterion ε=0.005 that can obtain trained neural network;
S43, by the input information of BP neural network, including 11 feature tags and user object as inspection data into
Row input, judges whether the result that neural network prediction model obtains is accurate.
S44, by optimizing to the continuous study of BP neural network, the accurate output valve of final output, that is, it is corresponding
The feature tag of identity, including existing feature tag, and generate unidentified feature tag.
Further, in the BP neural network, BP neural network uses the network topology of network structure 11 × 10 × 1
Structure, neuron function are Sigmoid characteristic functions.
Specific implementation mode
Case 1 is embodied:
A kind of user based on call center data draws a portrait method, using with collection industry, then its include the following steps:
S11 first by the time with smaller spaced discrete, and records all calls in the call observation phase;
S12 is acquired the sound recordings and fileinfo of call center, records and identifies to the collection in calling system
Word is changed into, word semanteme, including part-of-speech tagging, name Entity recognition, user-oriented dictionary function are extracted by Chinese word segmentation system
Deng, and extract user information and semantic information;
S13 arranges user information and semantic information, distinguishes user information, contextual data information and user's feelings
Thread information.
S21 clears up redundancy, such as duplicate message, without semantic information;
S23, the regularity then occurred to data are analyzed, and abnormal data is reasonably cleared data or change;
S3:The data cleaned to step S23 extract keyword, and such as " out of funds also not ", " rich not also " " are put and relied not
Also ", " no income ", " phone number usage time ", " arrearage situation ", " subjective attitude ", " user emotion " etc., and generate mark
Label, for summarizing the data such as individual character related to user, feature.
S41 carries out feature information extraction first, from state, call mesh and demand point, air time, the duration of call, visitor
Family place, IO circuits, number field, number information, the connection frequency, operation system information, client identity etc., satisfaction, noun
Category information, keyword message:It such as swears at people, complain, choosing 11 in quality inspection result, voice, intonation as feature tag, together
When choose and carried out questionnaire survey, and marked the user of identity as training object, by 11 feature tags and user
Input information of the object as BP neural network;
S42, by the input information of BP neural network, including 11 feature tags and user object as training data and
11 feature tags and user object are input to BP neural network, for training by inspection data when as training data
Neural network takes learning rate η=0.3, error criterion ε=0.005 that can obtain trained neural network;
S43, by the input information of BP neural network, including 11 feature tags and user object as inspection data into
Row input, judges whether the result that neural network prediction model obtains is accurate.
S44, by optimizing to the continuous study of BP neural network, the accurate output valve of final output, that is, it is corresponding
The feature tag of identity, including existing feature tag, and generate unidentified feature tag.
S5:Determine that the Predict Test of corresponding user and words art are incomplete to supplement according to the analysis result in step S44
User information data;
S6:The user information data in extraction supplement step S5 is excavated by communication data.
Several embodiments of the invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
Cannot the limitation to the scope of the claims of the present invention therefore be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the guarantor of the present invention
Protect range.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (7)
- A kind of method 1. user based on call center data draws a portrait comprising following steps:S1:By changing into word to recording, and semantic data is extracted by NLPIR Chinese word segmentation systems;S2:The data acquired in step S1 are cleaned;S3:Labeling processing is extracted to the step S2 data cleaned;S4:Clustering is carried out to the user tag in step S3, forms user's portrait;S5:Predict Test and the words art of corresponding user is determined according to the analysis result in step S4 to supplement incomplete user's letter Breath data;S6:The user information data in extraction supplement step S5 is excavated by communication data.
- 2. user's portrait method based on call center data as described in claim 1, it is characterised in that:The step S1 It comprises the following specific steps that:S11 first by the time with smaller spaced discrete, and records all calls in the call observation phase;S12 is acquired the sound recordings and fileinfo of call center, and extracts user information and semantic information;S13 arranges user information and semantic information, distinguishes user information, contextual data information and user emotion letter Breath.
- 3. user's portrait method based on call center data as described in claim 1, it is characterised in that:The step S2 It comprises the following specific steps that:S21 carries out classification and scene analysis to collected data in S1 first;S22 then clears up redundancy;S23, the regularity then occurred to data are analyzed, and abnormal data is reasonably cleared data or change.
- 4. user's portrait method based on call center data as described in claim 1, it is characterised in that:The step S3 It comprises the following specific steps that:Keyword is extracted, and generates label, it is several for summarizing hobby related to user, individual character, feature etc. According to.
- 5. user's portrait method based on call center data as described in claim 1, it is characterised in that:The step S4 It comprises the following specific steps that:S41 carries out feature information extraction first, from state, call mesh and demand point, air time, the duration of call, client Point, IO circuits, number field, number information, the connection frequency, operation system information, client identity etc., satisfaction, name part of speech letter Breath, keyword message:It such as swears at people, complain, choosing 11 in quality inspection result, voice, intonation as feature tag, selecting simultaneously It takes and had carried out questionnaire survey, and marked the user of identity as training object, by 11 feature tags and user object Input information as BP neural network;S42, by the input information of BP neural network, including 11 feature tags and user object are as training data and inspection 11 feature tags and user object are input to BP neural network by data when as training data, for training nerve Network takes learning rate η=0.3, error criterion ε=0.005 that can obtain trained neural network;S43, by the input information of BP neural network, including 11 feature tags and user object carried out as inspection data it is defeated Enter, judges whether the result that neural network prediction model obtains is accurate.S44, by the continuous study optimization to BP neural network, the accurate output valve of final output, that is, corresponding identity Feature tag, including existing feature tag, and the unidentified feature tag that generates.
- 6. user's portrait method based on call center data as claimed in claim 5, it is characterised in that:It is used in step S4 To BP neural network, the formula of clustering is as follows in the BP neural network, enables P={ P1, P2..., PmIndicate user's portrait collection It closes, wherein PiIndicate i-th of user's portrait.Ci={ Pi1, Pi2..., PinIndicate i-th of label after cluster, wherein Pij, table Show CiJ-th of element in classification.After cluster, the condition that respectively classifying meet is Uk I=1Ci=PBased on cluster result, it can be found that hidden between complicated user's portrait The information contained extracts new user's portrait label.
- 7. user's portrait method based on call center data as claimed in claim 5, it is characterised in that:The BP nerves In network, it is Sigmoid features that BP neural network, which uses the network topology structure of network structure 11 × 10 × 1, neuron function, Function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810256594.0A CN108550050A (en) | 2018-03-27 | 2018-03-27 | A kind of user's portrait method based on call center data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810256594.0A CN108550050A (en) | 2018-03-27 | 2018-03-27 | A kind of user's portrait method based on call center data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108550050A true CN108550050A (en) | 2018-09-18 |
Family
ID=63517225
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810256594.0A Pending CN108550050A (en) | 2018-03-27 | 2018-03-27 | A kind of user's portrait method based on call center data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108550050A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109215654A (en) * | 2018-10-22 | 2019-01-15 | 北京智合大方科技有限公司 | The mobile terminal intelligent customer service auxiliary system of Real-time speech recognition and natural language processing |
CN109658923A (en) * | 2018-10-19 | 2019-04-19 | 平安科技(深圳)有限公司 | Voice quality detecting method, equipment, storage medium and device based on artificial intelligence |
CN109711892A (en) * | 2018-12-28 | 2019-05-03 | 浙江百应科技有限公司 | The method for automatically generating client's label during Intelligent voice dialog |
CN110717733A (en) * | 2019-10-09 | 2020-01-21 | 浙江百应科技有限公司 | Enterprise work mobile phone intelligent management technology solution |
CN111292733A (en) * | 2018-12-06 | 2020-06-16 | 阿里巴巴集团控股有限公司 | Voice interaction method and device |
CN111866292A (en) * | 2020-07-20 | 2020-10-30 | 普强信息技术(北京)有限公司 | Application method of customer label based on voice data |
CN113228606A (en) * | 2018-11-21 | 2021-08-06 | N3有限责任公司 | Semantic CRM copies from mobile communication sessions |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070143235A1 (en) * | 2005-12-21 | 2007-06-21 | International Business Machines Corporation | Method, system and computer program product for organizing data |
US8682883B2 (en) * | 2011-04-14 | 2014-03-25 | Predictix Llc | Systems and methods for identifying sets of similar products |
CN103811009A (en) * | 2014-03-13 | 2014-05-21 | 华东理工大学 | Smart phone customer service system based on speech analysis |
CN106651424A (en) * | 2016-09-28 | 2017-05-10 | 国网山东省电力公司电力科学研究院 | Electric power user figure establishment and analysis method based on big data technology |
CN106686191A (en) * | 2015-11-06 | 2017-05-17 | 北京奇虎科技有限公司 | Processing method for adaptively identifying harassing call and processing system thereof |
CN106709754A (en) * | 2016-11-25 | 2017-05-24 | 云南电网有限责任公司昆明供电局 | Power user grouping method based on text mining |
CN107203910A (en) * | 2017-05-27 | 2017-09-26 | 科技谷(厦门)信息技术有限公司 | A kind of big data intelligent accurate marketing system |
CN107293310A (en) * | 2017-06-28 | 2017-10-24 | 上海航动科技有限公司 | A kind of user emotion analysis method and system |
CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
CN107808307A (en) * | 2017-09-28 | 2018-03-16 | 平安科技(深圳)有限公司 | Business personnel's picture forming method, electronic installation and computer-readable recording medium |
-
2018
- 2018-03-27 CN CN201810256594.0A patent/CN108550050A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070143235A1 (en) * | 2005-12-21 | 2007-06-21 | International Business Machines Corporation | Method, system and computer program product for organizing data |
US8682883B2 (en) * | 2011-04-14 | 2014-03-25 | Predictix Llc | Systems and methods for identifying sets of similar products |
CN103811009A (en) * | 2014-03-13 | 2014-05-21 | 华东理工大学 | Smart phone customer service system based on speech analysis |
CN106686191A (en) * | 2015-11-06 | 2017-05-17 | 北京奇虎科技有限公司 | Processing method for adaptively identifying harassing call and processing system thereof |
CN106651424A (en) * | 2016-09-28 | 2017-05-10 | 国网山东省电力公司电力科学研究院 | Electric power user figure establishment and analysis method based on big data technology |
CN106709754A (en) * | 2016-11-25 | 2017-05-24 | 云南电网有限责任公司昆明供电局 | Power user grouping method based on text mining |
CN107203910A (en) * | 2017-05-27 | 2017-09-26 | 科技谷(厦门)信息技术有限公司 | A kind of big data intelligent accurate marketing system |
CN107293310A (en) * | 2017-06-28 | 2017-10-24 | 上海航动科技有限公司 | A kind of user emotion analysis method and system |
CN107808307A (en) * | 2017-09-28 | 2018-03-16 | 平安科技(深圳)有限公司 | Business personnel's picture forming method, electronic installation and computer-readable recording medium |
CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
Non-Patent Citations (1)
Title |
---|
赵守香等: "《大数据分析与应用》", 31 December 2015, 航空工业出版社 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109658923A (en) * | 2018-10-19 | 2019-04-19 | 平安科技(深圳)有限公司 | Voice quality detecting method, equipment, storage medium and device based on artificial intelligence |
CN109658923B (en) * | 2018-10-19 | 2024-01-30 | 平安科技(深圳)有限公司 | Speech quality inspection method, equipment, storage medium and device based on artificial intelligence |
CN109215654A (en) * | 2018-10-22 | 2019-01-15 | 北京智合大方科技有限公司 | The mobile terminal intelligent customer service auxiliary system of Real-time speech recognition and natural language processing |
CN113228606A (en) * | 2018-11-21 | 2021-08-06 | N3有限责任公司 | Semantic CRM copies from mobile communication sessions |
CN111292733A (en) * | 2018-12-06 | 2020-06-16 | 阿里巴巴集团控股有限公司 | Voice interaction method and device |
CN109711892A (en) * | 2018-12-28 | 2019-05-03 | 浙江百应科技有限公司 | The method for automatically generating client's label during Intelligent voice dialog |
CN110717733A (en) * | 2019-10-09 | 2020-01-21 | 浙江百应科技有限公司 | Enterprise work mobile phone intelligent management technology solution |
CN111866292A (en) * | 2020-07-20 | 2020-10-30 | 普强信息技术(北京)有限公司 | Application method of customer label based on voice data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108550050A (en) | A kind of user's portrait method based on call center data | |
US11631147B2 (en) | Social recommendation method based on multi-feature heterogeneous graph neural networks | |
CN109189901B (en) | Method for automatically discovering new classification and corresponding corpus in intelligent customer service system | |
CN108920544A (en) | A kind of personalized position recommended method of knowledge based map | |
CN109558429A (en) | The two-way recommendation method and system of talent service based on internet big data | |
CN105608477A (en) | Method and system for matching portraits with positions | |
CN105868317B (en) | Digital education resource recommendation method and system | |
CN107958091A (en) | A kind of NLP artificial intelligence approaches and interactive system based on financial vertical knowledge mapping | |
CN109767318A (en) | Loan product recommended method, device, equipment and storage medium | |
CN104142995B (en) | The social event recognition methods of view-based access control model attribute | |
CN111159387B (en) | Recommendation method based on multi-dimensional alarm information text similarity analysis | |
CA3052174A1 (en) | System and method for call centre management | |
CN106161209B (en) | A kind of method for filtering spam short messages and system based on depth self study | |
CN109543034A (en) | Text Clustering Method, device and the readable storage medium storing program for executing of knowledge based map | |
CN111049818B (en) | Abnormal information discovery method based on network traffic big data | |
CN108897778A (en) | A kind of image labeling method based on multi-source big data analysis | |
CN107193974A (en) | Localized information based on artificial intelligence determines method and apparatus | |
CN109918648B (en) | Rumor depth detection method based on dynamic sliding window feature score | |
CN112235470B (en) | Incoming call client follow-up method, device and equipment based on voice recognition | |
CN107491729A (en) | The Handwritten Digit Recognition method of convolutional neural networks based on cosine similarity activation | |
Dong et al. | Insect sound recognition based on convolutional neural network | |
CN107679075A (en) | Method for monitoring network and equipment | |
CN110727864A (en) | User portrait method based on mobile phone App installation list | |
CN104657466A (en) | Method and device for identifying user interest based on forum post features | |
CN107368499A (en) | A kind of client's tag modeling and recommendation method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180918 |