CN104361507A - Commodity recommending method and system - Google Patents
Commodity recommending method and system Download PDFInfo
- Publication number
- CN104361507A CN104361507A CN201410668170.7A CN201410668170A CN104361507A CN 104361507 A CN104361507 A CN 104361507A CN 201410668170 A CN201410668170 A CN 201410668170A CN 104361507 A CN104361507 A CN 104361507A
- Authority
- CN
- China
- Prior art keywords
- user
- commodity
- targeted customer
- text
- database
- 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
Abstract
The invention discloses a commodity recommending method and system. The commodity recommending method comprises the following steps that voice information input by a target user is received; the voice information is recognized and converted into text information; the text information is processed to generate structured text; inquiring is conducted to obtain a voice inquiring result; the structured text is recorded into a log of the target user; clustering analysis is conducted, and a user cluster which the target user belongs to is obtained; the user closest to the target user is searched for; potential grades, given by the target user, of commodities are calculated according to the average grades, given by the user closet to the target user, of the commodities and the average grades of the target user, and the commodities are screened according to the potential grades to generate a commodity set; a plurality of commodities are selected from the commodity set and recommended to the target user according to the structured text. The commodity recommending method and system avoid a large amount of irrelevant information generated when the user searches for needed commodities in the browsing process in e-commerce, greatly improve the efficiency of the e-commerce and meanwhile greatly improve the use experience of the user.
Description
Technical field
The present invention relates to a kind of Method of Commodity Recommendation and system.
Background technology
Along with the continuous expansion of ecommerce scale, commodity number and kind increase fast, and customer need spends a large amount of time just can find the commodity oneself wanting to buy.Thisly browse a large amount of irrelevant information and product process can make the consumer be submerged in problem of information overload constantly run off undoubtedly.
On the other hand, voice query function has very actual application in every profession and trade.Along with popularizing of mobile terminal, increasing speech recognition application is developed.Utilize speech recognition technology can be convenient to search or the operation of user to a certain extent, initiatively carry out screening the system of recommending to commodity according to user's request if speech recognition technology can be aided with, then be expected to sooner, conveniently recommend the product of the most applicable user, to reach the object making user can look for required commodity more easily, thus greatly improve the experience of user.
Summary of the invention
The technical problem to be solved in the present invention to require a great deal of time the commodity that just can find and oneself want to buy in order to user in existing ecommerce, and in this browsing, certainly lead to a large amount of irrelevant information, thus the efficiency of ecommerce can greatly be reduced, have a strong impact on the defect of the experience of user simultaneously, propose a kind of Method of Commodity Recommendation and system.
The present invention solves above-mentioned technical matters by following technical proposals:
The invention provides a kind of Method of Commodity Recommendation, its feature is, in a database, record multiple user journals of multiple user, user journal records the historical record of user correspondingly, and this Method of Commodity Recommendation comprises the following steps:
S
1, receive the voice messaging of user's input, and using this user as targeted customer;
S
2, to carry out speech recognition conversion to voice messaging be a text message;
S
3, structuring process is carried out to text information, to generate a structured text;
S
4, inquire about in a database according to structured text, to obtain a speech polling result;
S
5if, in this database, record the user journal of this targeted customer, then this structured text is recorded in the user journal of this targeted customer, if do not record the user journal of this targeted customer in this database, then in this database, create the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer;
S
6, according to the user journal of the plurality of user recorded in this database, cluster analysis is carried out to user, to obtain cluster result, this cluster result comprises some user clusterings and feature thereof;
S
7, determine belonging to this targeted customer user clustering;
S
8, according to Collaborative Filtering Recommendation Algorithm, in the user clustering of correspondence, search for the arest neighbors user of this targeted customer;
S
9, according to this arest neighbors user, the potential scoring of this targeted customer to commodity is calculated to the scoring of commodity and the average score of this targeted customer, according to potential scoring screening commodity to generate candidate's commodity collection;
S
10, according to this structured text from these candidate's commodity concentrate select some commodity, and by this some commercial product recommending give this targeted customer.
Wherein user journal can be the history speech polling daily record of user, have collected the contents such as the historical behavior of user, for the basis as subsequent analysis process such as cluster analysis, neighbour user determine in user journal.
Above-mentioned steps S
9for setting up the model for analyzing user preferences according to user journal, according to the behavior record of the user profile of collecting, the information of user can be extracted according to multiple dimension, sets up the model of user preferences or preference.On the basis of user preferences or preference, utilize the historical information of user more adequately target of prediction user degree is liked to particular commodity, and according to this fancy grade, Recommendations are carried out to targeted customer.
Preferably, this step S
2comprise the following steps:
S
21, voice messaging is carried out to the extraction of audio frequency characteristics;
S
22, obtain the acoustic model set up based on Markov model, then resolve according to this acoustic model and obtain and word sequence that the audio frequency characteristics matching degree extracted is the highest;
S
23, this word sequence is generated as text information.
In above-mentioned steps, acoustic model is set up based on Markov model, and set up comprise system institute treatable word finder and its pronounce pronunciation model (can be contained in acoustic model), in pronunciation model, contain the mapping between acoustic model modeling unit and language model modeling unit.According to acoustic model marking, find a word Model sequence to describe input speech signal, thus obtain corresponding word decoding sequence.Said process also can be regarded as, and to the signal of input, according to acoustic model and dictionary, looks for word string or the word sequence of maximum probability output.
Preferably, this step S
3comprise the following steps:
S
31, utilize preset participle and part-of-speech tagging algorithm participle and part-of-speech tagging are carried out to text information;
S
32, synonym replacement is carried out, to be normalized to the text information after participle;
S
33, find out the first kind word that text information comprises, and replace the first kind word in text information, to form this structured text with the asterisk wildcard preset.
Preferably, this step S
21for: calculating voice messaging being carried out to analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics.LPC is wherein Linear Predictive Coding, is translated into linear predictive coding, encodes also referred to as sound source.
Preferably, this step S
7for: calculate user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to mark both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user.
Through above-mentioned steps S
7the cluster result that formed of process, be in fact substantially user commodity to akin point of interest is put into same cluster.
Present invention also offers a kind of commercial product recommending system, it comprises:
Database, wherein record multiple user journals of multiple user, user journal records the historical record of user correspondingly;
Load module, for receiving the voice messaging of user's input, and using this user as targeted customer;
Sound identification module is a text message for carrying out speech recognition conversion to voice messaging;
Structurized module, for carrying out structuring process to text information, to generate a structured text;
Enquiry module, for inquiring about in a database according to structured text, to obtain a speech polling result;
Logger module, for recording the user journal of this targeted customer in this database, this structured text being recorded in the user journal of this targeted customer, in this database, creating the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer when not recording the user journal of this targeted customer in this database;
Cluster Analysis module, for carrying out cluster analysis according to the user journal of the plurality of user recorded in this database to user, to obtain cluster result, this cluster result comprises some user clusterings and feature thereof;
Belong to class determination module, for determining the user clustering belonging to this targeted customer;
Neighbor search module, for according to Collaborative Filtering Recommendation Algorithm, searches for the arest neighbors user of this targeted customer in the user clustering of correspondence;
Screening module, for calculating the potential scoring of this targeted customer to commodity according to this arest neighbors user to the scoring of commodity and the average score of this targeted customer, according to potential scoring screening commodity to generate candidate's commodity collection;
Recommending module, for selecting some commodity according to this structured text from these candidate's commodity are concentrated, and gives this targeted customer by this some commercial product recommending.
Preferably, sound identification module comprises audio feature extraction unit, word sequence matching unit and text generation unit.
Audio feature extraction unit is used for extraction voice messaging being carried out to audio frequency characteristics.Word sequence matching unit, for obtaining the acoustic model set up based on Markov model, is then resolved according to this acoustic model and is obtained and word sequence that the audio frequency characteristics matching degree extracted is the highest.Text generation unit is used for this word sequence to be generated as text information.
Preferably, structurized module comprises part-of-speech tagging unit, synonym replacement unit, asterisk wildcard unit.
Part-of-speech tagging unit carries out participle and part-of-speech tagging for utilizing default participle and part-of-speech tagging algorithm to text information.Synonym replacement unit is used for carrying out synonym replacement, to be normalized to the text information after participle.The first kind word that asterisk wildcard unit comprises for finding out text information, and adopt the asterisk wildcard preset to replace first kind word in text information to form this structured text.What wherein relate to should be understood to prestore in systems in which for synonym, asterisk wildcard and its Substitution Rules of replacing.
Preferably, audio feature extraction unit is used for calculating voice messaging being carried out to analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics.
Preferably, belong to class determination module for calculating user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to mark both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user.
On the basis meeting this area general knowledge, above-mentioned each optimum condition, can combination in any, obtains the preferred embodiments of the invention.
Positive progressive effect of the present invention is:
A large amount of irrelevant informations during what Method of Commodity Recommendation of the present invention and system avoided that commodity needed for user search in ecommerce produce browse, substantially increase the efficiency of ecommerce, substantially improve the experience of user simultaneously.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the Method of Commodity Recommendation of the embodiment of the present invention 1.
Fig. 2 is the schematic diagram of the commercial product recommending system of the embodiment of the present invention 2.
Embodiment
Provide present pre-ferred embodiments below in conjunction with accompanying drawing, to describe technical scheme of the present invention in detail, but therefore do not limit the present invention among described scope of embodiments.
Embodiment 1
The Method of Commodity Recommendation of the present embodiment, in a database, record multiple user journals of multiple user, user journal records the historical record of user correspondingly.Shown in figure 1, the Method of Commodity Recommendation of the present embodiment comprises the following steps:
S
1, receive the voice messaging of user's input, and using this user as targeted customer;
S
2, to carry out speech recognition conversion to voice messaging be a text message;
S
3, structuring process is carried out to text information, to generate a structured text;
S
4, inquire about in a database according to structured text, to obtain a speech polling result;
S
5if, in this database, record the user journal of this targeted customer, then this structured text is recorded in the user journal of this targeted customer, if do not record the user journal of this targeted customer in this database, then in this database, create the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer;
S
6, according to the user journal of the plurality of user recorded in this database, cluster analysis is carried out to user, to obtain cluster result, this cluster result comprises some user clusterings and feature thereof;
S
7, calculate user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to mark both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user;
S
8, according to Collaborative Filtering Recommendation Algorithm, in the user clustering of correspondence, search for the arest neighbors user of this targeted customer;
S
9, according to this arest neighbors user, the potential scoring of this targeted customer to commodity is calculated to the scoring of commodity and the average score of this targeted customer, according to potential scoring screening commodity to generate candidate's commodity collection;
S
10, according to this structured text from these candidate's commodity concentrate select some commodity, and by this some commercial product recommending give this targeted customer.
Wherein, this step S
9for setting up the model for analyzing user preferences according to user journal, according to the behavior record of the user profile of collecting, extracting the information of user according to multiple dimension, setting up the model of user preferences or preference.On the basis of user preferences or preference, utilize the historical information of user more adequately target of prediction user degree is liked to particular commodity, and according to this fancy grade, Recommendations are carried out to targeted customer.
Step S specifically
2with step S
3comprise concrete steps as described below.
Step S
2comprise the following steps:
S
21, voice messaging carried out to the calculating of analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics;
S
22, obtain the acoustic model set up based on Markov model, then resolve according to this acoustic model and obtain and word sequence that the audio frequency characteristics matching degree extracted is the highest;
S
23, this word sequence is generated as text information.
Step S
3comprise the following steps:
S
31, utilize preset participle and part-of-speech tagging algorithm participle and part-of-speech tagging are carried out to text information;
S
32, synonym replacement is carried out, to be normalized to the text information after participle;
S
33, find out the first kind word that text information comprises, and replace the first kind word in text information, to form this structured text with the asterisk wildcard preset.
Embodiment 2
As shown in Figure 2, the commercial product recommending system of the present embodiment comprises: database 1, load module 2, sound identification module 3, structurized module 4, enquiry module 5, logger module 6, Cluster Analysis module 7, genus class determination module 8, neighbor search module 9, screening module 10 and recommending module 11.
Record multiple user journals of multiple user in database 1, user journal records the historical record of user correspondingly.The voice messaging that load module 2 inputs for receiving a user, and using this user as targeted customer.Sound identification module 3 is a text message for carrying out speech recognition conversion to voice messaging.
Structurized module 4 for carrying out structuring process to text information, to generate a structured text.Enquiry module 5 for inquiring about in a database according to structured text, to obtain a speech polling result.
Logger module 6 for recording the user journal of this targeted customer in this database, this structured text being recorded in the user journal of this targeted customer, in this database, creating the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer when not recording the user journal of this targeted customer in this database.
Cluster Analysis module 7 is for carrying out cluster analysis according to the user journal of the plurality of user recorded in this database to user, and to obtain cluster result, this cluster result comprises some user clusterings and feature thereof.Belong to class determination module 8 for determining the user clustering belonging to this targeted customer.Neighbor search module 9, for according to Collaborative Filtering Recommendation Algorithm, searches for the arest neighbors user of this targeted customer in the user clustering of correspondence;
Screening module 10, for calculating the potential scoring of this targeted customer to commodity according to this arest neighbors user to the scoring of commodity and the average score of this targeted customer, screens commodity to generate candidate's commodity collection according to potential scoring.This some commercial product recommending for selecting some commodity according to this structured text from these candidate's commodity are concentrated, and is given this targeted customer by recommending module 11.
Wherein, sound identification module comprises audio feature extraction unit, word sequence matching unit and text generation unit.And structurized module comprises part-of-speech tagging unit, synonym replacement unit, asterisk wildcard unit.
Specifically, audio feature extraction unit is used for calculating voice messaging being carried out to analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics.
Word sequence matching unit, for obtaining the acoustic model set up based on Markov model, is then resolved according to this acoustic model and is obtained and word sequence that the audio frequency characteristics matching degree extracted is the highest.Text generation unit is used for this word sequence to be generated as text information.
Part-of-speech tagging unit carries out participle and part-of-speech tagging for utilizing default participle and part-of-speech tagging algorithm to text information.Synonym replacement unit is used for carrying out synonym replacement, to be normalized to the text information after participle.The first kind word that asterisk wildcard unit comprises for finding out text information, and adopt the asterisk wildcard preset to replace first kind word in text information to form this structured text.What wherein relate to should be understood to prestore in systems in which for synonym, asterisk wildcard and its Substitution Rules of replacing.
Belong to class determination module for calculating user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to mark both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user.
Although the foregoing describe the specific embodiment of the present invention, it will be understood by those of skill in the art that these only illustrate, protection scope of the present invention is defined by the appended claims.Those skilled in the art, under the prerequisite not deviating from principle of the present invention and essence, can make various changes or modifications to these embodiments, but these change and amendment all falls into protection scope of the present invention.
Claims (10)
1. a Method of Commodity Recommendation, is characterized in that, in a database, record multiple user journals of multiple user, user journal records the historical record of user correspondingly, and this Method of Commodity Recommendation comprises the following steps:
S
1, receive the voice messaging of user's input, and using this user as targeted customer;
S
2, to carry out speech recognition conversion to voice messaging be a text message;
S
3, structuring process is carried out to text information, to generate a structured text;
S
4, inquire about in a database according to structured text, to obtain a speech polling result;
S
5if, in this database, record the user journal of this targeted customer, then this structured text is recorded in the user journal of this targeted customer, if do not record the user journal of this targeted customer in this database, then in this database, create the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer;
S
6, according to the user journal of the plurality of user recorded in this database, cluster analysis is carried out to user, to obtain cluster result, this cluster result comprises some user clusterings and feature thereof;
S
7, determine belonging to this targeted customer user clustering;
S
8, according to Collaborative Filtering Recommendation Algorithm, in the user clustering of correspondence, search for the arest neighbors user of this targeted customer;
S
9, according to this arest neighbors user, the potential scoring of this targeted customer to commodity is calculated to the scoring of commodity and the average score of this targeted customer, according to potential scoring screening commodity to generate candidate's commodity collection;
S
10, according to this structured text from these candidate's commodity concentrate select some commodity, and by this some commercial product recommending give this targeted customer.
2. Method of Commodity Recommendation as claimed in claim 1, is characterized in that, this step S
2comprise the following steps:
S
21, voice messaging is carried out to the extraction of audio frequency characteristics;
S
22, obtain the acoustic model set up based on Markov model, then resolve according to this acoustic model and obtain and word sequence that the audio frequency characteristics matching degree extracted is the highest;
S
23, this word sequence is generated as text information.
3. Method of Commodity Recommendation as claimed in claim 1, is characterized in that, this step S
3comprise the following steps:
S
31, utilize preset participle and part-of-speech tagging algorithm participle and part-of-speech tagging are carried out to text information;
S
32, synonym replacement is carried out, to be normalized to the text information after participle;
S
33, find out the first kind word that text information comprises, and replace the first kind word in text information, to form this structured text with the asterisk wildcard preset.
4. Method of Commodity Recommendation as claimed in claim 2, is characterized in that, this step S
21for: calculating voice messaging being carried out to analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics.
5. as the Method of Commodity Recommendation in claim 1-4 as described in any one, it is characterized in that, this step S
7for: calculate user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to mark both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user.
6. a commercial product recommending system, is characterized in that, comprising:
Database, wherein record multiple user journals of multiple user, user journal records the historical record of user correspondingly;
Load module, for receiving the voice messaging of user's input, and using this user as targeted customer;
Sound identification module is a text message for carrying out speech recognition conversion to voice messaging;
Structurized module, for carrying out structuring process to text information, to generate a structured text;
Enquiry module, for inquiring about in a database according to structured text, to obtain a speech polling result;
Logger module, for recording the user journal of this targeted customer in this database, this structured text being recorded in the user journal of this targeted customer, in this database, creating the user journal of this targeted customer and this structured text is recorded in the user journal of this targeted customer when not recording the user journal of this targeted customer in this database;
Cluster Analysis module, for carrying out cluster analysis according to the user journal of the plurality of user recorded in this database to user, to obtain cluster result, this cluster result comprises some user clusterings and feature thereof;
Belong to class determination module, for determining the user clustering belonging to this targeted customer;
Neighbor search module, for according to Collaborative Filtering Recommendation Algorithm, searches for the arest neighbors user of this targeted customer in the user clustering of correspondence;
Screening module, for calculating the potential scoring of this targeted customer to commodity according to this arest neighbors user to the scoring of commodity and the average score of this targeted customer, according to potential scoring screening commodity to generate candidate's commodity collection;
Recommending module, for selecting some commodity according to this structured text from these candidate's commodity are concentrated, and gives this targeted customer by this some commercial product recommending.
7. commercial product recommending system as claimed in claim 6, it is characterized in that, sound identification module comprises audio feature extraction unit, word sequence matching unit and text generation unit, audio feature extraction unit is used for extraction voice messaging being carried out to audio frequency characteristics, word sequence matching unit for obtain set up based on Markov model acoustic model, then resolve according to this acoustic model and obtain and word sequence that the audio frequency characteristics matching degree extracted is the highest, text generation unit is used for this word sequence to be generated as text information.
8. commercial product recommending system as claimed in claim 6, it is characterized in that, structurized module comprises part-of-speech tagging unit, synonym replacement unit, asterisk wildcard unit, part-of-speech tagging unit carries out participle and part-of-speech tagging for utilizing default participle and part-of-speech tagging algorithm to text information, synonym replacement unit is used for carrying out synonym replacement to the text information after participle, to be normalized, the first kind word that asterisk wildcard unit comprises for finding out text information, and with the first kind word in the asterisk wildcard preset replacement text information to form this structured text.
9. commercial product recommending system as claimed in claim 7, it is characterized in that, audio feature extraction unit is used for calculating voice messaging being carried out to analog to digital conversion, end-point detection, pre-emphasis, windowing, autocorrelation sequence, LPC coefficient and/or cepstral coefficients, to extract audio frequency characteristics.
10. as the commercial product recommending system in claim 6-9 as described in any one, it is characterized in that, belong to class determination module for calculating user to the concern of commodity and user to the similarity of the scoring of commodity, and will the property paid close attention to and scoring both linear combination to form clustering analysis parameter, then utilize this cluster analysis parameter to carry out cluster to user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410668170.7A CN104361507A (en) | 2014-11-20 | 2014-11-20 | Commodity recommending method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410668170.7A CN104361507A (en) | 2014-11-20 | 2014-11-20 | Commodity recommending method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104361507A true CN104361507A (en) | 2015-02-18 |
Family
ID=52528765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410668170.7A Pending CN104361507A (en) | 2014-11-20 | 2014-11-20 | Commodity recommending method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104361507A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104618819A (en) * | 2015-03-05 | 2015-05-13 | 广州新节奏智能科技有限公司 | Television terminal-based 3D somatosensory shopping system and method |
| CN105574173A (en) * | 2015-12-18 | 2016-05-11 | 畅捷通信息技术股份有限公司 | Commodity searching method and commodity searching device based on voice recognition |
| CN105653738A (en) * | 2016-03-01 | 2016-06-08 | 北京百度网讯科技有限公司 | Search result broadcasting method and device based on artificial intelligence |
| CN105787770A (en) * | 2016-04-27 | 2016-07-20 | 上海遥薇(集团)有限公司 | Non-negative matrix factorization (NMF) algorithm-based big data commodity and service recommending method and system |
| CN105824942A (en) * | 2016-03-21 | 2016-08-03 | 上海珍岛信息技术有限公司 | Item recommendation method and system based on collaborative filtering algorithm |
| CN106570020A (en) * | 2015-10-09 | 2017-04-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus used for providing recommended information |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
| CN107886949A (en) * | 2017-11-24 | 2018-04-06 | 科大讯飞股份有限公司 | A kind of content recommendation method and device |
| CN107967641A (en) * | 2017-10-18 | 2018-04-27 | 美的智慧家居科技有限公司 | Method of Commodity Recommendation, device and computer-readable recording medium |
| CN107993134A (en) * | 2018-01-23 | 2018-05-04 | 北京知行信科技有限公司 | A kind of smart shopper exchange method and system based on user interest |
| CN107993133A (en) * | 2018-01-23 | 2018-05-04 | 北京知行信科技有限公司 | A kind of intellectual analysis based on natural language recommends method and system |
| CN109087110A (en) * | 2018-08-24 | 2018-12-25 | 深圳市云之音科技有限公司 | A kind of intelligence marketing method, device and terminal device |
| WO2019149132A1 (en) * | 2018-02-01 | 2019-08-08 | 腾讯科技（深圳）有限公司 | Audio information processing method, device, storage medium and electronic device |
| CN112070519A (en) * | 2019-06-11 | 2020-12-11 | 中国科学院沈阳自动化研究所 | Prediction method based on data global search and feature classification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231660A (en) * | 2008-02-19 | 2008-07-30 | 林超 | System and method for digging key information of telephony nature conversation |
| CN101539906A (en) * | 2008-03-17 | 2009-09-23 | 亿维讯软件（北京）有限公司 | System and method for automatically analyzing patent text |
| CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommending method and system which integrate attributes and structural similarity |
| CN102915341A (en) * | 2012-09-21 | 2013-02-06 | 人民搜索网络股份公司 | Dynamic topic model-based dynamic text cluster device and method |
| CN103793515A (en) * | 2014-02-11 | 2014-05-14 | 安徽科大讯飞信息科技股份有限公司 | Service voice intelligent search and analysis system and method |
-
2014
- 2014-11-20 CN CN201410668170.7A patent/CN104361507A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231660A (en) * | 2008-02-19 | 2008-07-30 | 林超 | System and method for digging key information of telephony nature conversation |
| CN101539906A (en) * | 2008-03-17 | 2009-09-23 | 亿维讯软件（北京）有限公司 | System and method for automatically analyzing patent text |
| CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommending method and system which integrate attributes and structural similarity |
| CN102915341A (en) * | 2012-09-21 | 2013-02-06 | 人民搜索网络股份公司 | Dynamic topic model-based dynamic text cluster device and method |
| CN103793515A (en) * | 2014-02-11 | 2014-05-14 | 安徽科大讯飞信息科技股份有限公司 | Service voice intelligent search and analysis system and method |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104618819A (en) * | 2015-03-05 | 2015-05-13 | 广州新节奏智能科技有限公司 | Television terminal-based 3D somatosensory shopping system and method |
| CN106570020A (en) * | 2015-10-09 | 2017-04-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus used for providing recommended information |
| CN105574173A (en) * | 2015-12-18 | 2016-05-11 | 畅捷通信息技术股份有限公司 | Commodity searching method and commodity searching device based on voice recognition |
| CN105653738B (en) * | 2016-03-01 | 2020-05-22 | 北京百度网讯科技有限公司 | Search result broadcasting method and device based on artificial intelligence |
| CN105653738A (en) * | 2016-03-01 | 2016-06-08 | 北京百度网讯科技有限公司 | Search result broadcasting method and device based on artificial intelligence |
| US10810272B2 (en) | 2016-03-01 | 2020-10-20 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for broadcasting search result based on artificial intelligence |
| CN105824942A (en) * | 2016-03-21 | 2016-08-03 | 上海珍岛信息技术有限公司 | Item recommendation method and system based on collaborative filtering algorithm |
| CN105787770A (en) * | 2016-04-27 | 2016-07-20 | 上海遥薇(集团)有限公司 | Non-negative matrix factorization (NMF) algorithm-based big data commodity and service recommending method and system |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
| CN107967641A (en) * | 2017-10-18 | 2018-04-27 | 美的智慧家居科技有限公司 | Method of Commodity Recommendation, device and computer-readable recording medium |
| CN107886949A (en) * | 2017-11-24 | 2018-04-06 | 科大讯飞股份有限公司 | A kind of content recommendation method and device |
| CN107993133A (en) * | 2018-01-23 | 2018-05-04 | 北京知行信科技有限公司 | A kind of intellectual analysis based on natural language recommends method and system |
| CN107993134A (en) * | 2018-01-23 | 2018-05-04 | 北京知行信科技有限公司 | A kind of smart shopper exchange method and system based on user interest |
| WO2019149132A1 (en) * | 2018-02-01 | 2019-08-08 | 腾讯科技（深圳）有限公司 | Audio information processing method, device, storage medium and electronic device |
| US11475894B2 (en) | 2018-02-01 | 2022-10-18 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for providing feedback information based on audio input |
| CN109087110A (en) * | 2018-08-24 | 2018-12-25 | 深圳市云之音科技有限公司 | A kind of intelligence marketing method, device and terminal device |
| CN112070519A (en) * | 2019-06-11 | 2020-12-11 | 中国科学院沈阳自动化研究所 | Prediction method based on data global search and feature classification |
| CN112070519B (en) * | 2019-06-11 | 2024-03-05 | 中国科学院沈阳自动化研究所 | Prediction method based on data global search and feature classification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104361507A (en) | Commodity recommending method and system | |
| CN105787770A (en) | Non-negative matrix factorization (NMF) algorithm-based big data commodity and service recommending method and system | |
| JP6247292B2 (en) | Query expansion | |
| US10387805B2 (en) | System and method for ranking news feeds | |
| CN107544988B (en) | Method and device for acquiring public opinion data | |
| CN104462336A (en) | Information pushing method and device | |
| CN104751354B (en) | A kind of advertisement crowd screening technique | |
| CN105095210A (en) | Method and apparatus for screening promotional keywords | |
| KR102361597B1 (en) | A program recording medium on which a program for labeling sentiment information in news articles using big data is recoded | |
| CN101833560A (en) | Manufacturer public praise automatic sequencing system based on internet | |
| CN105468649B (en) | Method and device for judging matching of objects to be displayed | |
| KR101804967B1 (en) | Method and system to recommend music contents by database composed of user's context, recommended music and use pattern | |
| CN111832290A (en) | Model training method and device for determining text relevancy, electronic equipment and readable storage medium | |
| CN109063171B (en) | Resource matching method based on semantics | |
| CN110246496A (en) | Audio recognition method, system, computer equipment and storage medium | |
| CN103309869A (en) | Method and system for recommending display keyword of data object | |
| KR20200053005A (en) | Method for provide enterprise stock forecasting algorithm based on situation cognition | |
| CN111260428A (en) | Commodity recommendation method and device | |
| CN106372956B (en) | Method and system for identifying intention entity based on user search log | |
| Lee et al. | A hybrid collaborative filtering-based product recommender system using search keywords | |
| CN113282754A (en) | Public opinion detection method, device, equipment and storage medium for news events | |
| CN113407729B (en) | Judicial-oriented personalized case recommendation method and system | |
| CN113988195A (en) | Private domain traffic clue mining method and device, vehicle and readable medium | |
| KR102358357B1 (en) | Estimating apparatus for market size, and control method thereof | |
| CN105138552A (en) | Fashion tendency analysis system mining online sale data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160206 Address after: 200335 Shanghai city Changning District Admiralty Road No. 968 Building No. 16 10 floor Applicant after: SHANGHAI XIECHENG BUSINESS CO., LTD. Address before: 200335 Shanghai City, Changning District Fuquan Road No. 99, Ctrip network technology building Applicant before: Ctrip computer technology (Shanghai) Co., Ltd. |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150218 |
|
| RJ01 | Rejection of invention patent application after publication |