CN109785052A - Smart shopper method and system based on dark data mining - Google Patents
Smart shopper method and system based on dark data mining Download PDFInfo
- Publication number
- CN109785052A CN109785052A CN201811598887.3A CN201811598887A CN109785052A CN 109785052 A CN109785052 A CN 109785052A CN 201811598887 A CN201811598887 A CN 201811598887A CN 109785052 A CN109785052 A CN 109785052A
- Authority
- CN
- China
- Prior art keywords
- information
- audio
- keyword
- video
- server
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to dark data mining applied technical field more particularly to a kind of smart shopper method and system based on dark data mining；System includes: audio-video capture devices, for obtaining the current time axis information in currently playing audio/video information, and sends it to information server and is handled；Information server, for according to the current time axis information received, corresponding keyword is extracted in keyword data, and goods matching is carried out in merchandising database according to the keyword extracted and recommends to obtain corresponding Recommendations information, it is then sent to commercial product recommending terminal and carries out display output；And commercial product recommending terminal.Smart shopper system disclosed in this invention based on dark data mining, can the covering of high efficiency height at low cost using in video in the market, by its dark data " lighting ", realize that the intelligence of shopping information is promoted.
Description
Technical field
The present invention relates to dark data mining applied technical field more particularly to a kind of smart shoppers based on dark data mining
Method and system.
Background technique
Natural language processing (NLP) is computer science, artificial intelligence and philological crossing domain.Purpose is to allow calculating
Machine processing or " understanding " natural language, to execute the tasks such as language translation and question answering.With speech interface and chat
The rise of robot, NLP become one of information age most important technology, are the important components of artificial intelligence.The mankind
Language be mostly discrete/symbol/classification signal system, probably because in this way signal reliability it is higher.A kind of language
Class symbol can be encoded in several ways signal of communication: sound, gesture, write, image etc..Human language can use it
Middle any one signal indicates.Human language is indefinite (different from programming and other formal languages).Therefore, to mankind's language
Expression, study and the complexity using language/situation/context/vocabulary/visual knowledge with height of speech.
Dark data refer to those data for not being exploited or understanding, from the dark side of big data.It includes not yet answering
For business, forms competitive information or generate all data objects and type of help to business decision.Dark data point
The initial data primarily directed to textual form is analysed, such as text information, file, Email, audio video file and static state
Image etc. also can carry out dark data analysis without being searched the network information of engine index to deep net sometimes.It is most dark
Data are all classified among the poor information caused by structure, moreover, these data are only previously used once, finally, just being stored up
There are logical miss, among other community of substantial amounts.At present in social media, these internets of video website ...
In carrier, video has been used as main relay medium, but in the information for being included to video dark data i.e. described above,
It is not used by real excavate, wherein often containing huge information in the sound dimension of video, so to acoustic application
Natural language processing data are an excellent methods.
Therefore, to solve the above-mentioned problems, be badly in need of inventing a kind of new smart shopper method based on dark data mining and
System.
Summary of the invention
It is an object of the invention to: a kind of smart shopper method and system based on dark data mining are provided, high efficiency is high
Covering using in video in the market, by its dark data " lighting ", realizes that the intelligence of shopping information is promoted at low cost.
The present invention provides following proposals:
A kind of smart shopper method based on dark data mining, comprising the following steps:
Obtain video information and audio-frequency information；
The video information and audio-frequency information that will acquire are converted into the voice messaging to be identified of standard；
Voice messaging to be identified and acoustic model are matched to obtain meaning of one's words text；
By the meaning of one's words text received, temporally axis information classification storage is corresponding with video information and audio-frequency information difference
Keyword；
Keyword is constituted into keyword database；
According to the timeline information of current video information and audio-frequency information, corresponding keyword is extracted in keyword data；
Goods matching is carried out in merchandising database according to the keyword extracted to recommend to obtain corresponding Recommendations information simultaneously
Carry out display output.
Preferably, the tonic train resolved into the voice messaging to be identified of standard including video data, and the audio
Comprising the timeline information with audio sync in sequence.
Further, the present invention also provides the intelligence of a kind of smart shopper method described in realize based on dark data mining
Purchase system, comprising:
Audio-video capture devices, for obtaining video information and audio-frequency information, and the video information and audio-frequency information that will acquire are pre-
Processing server；Meanwhile the current time axis information in currently playing audio/video information is obtained, and send it to information clothes
Business device is handled；
Preprocessing server, video information and audio-frequency information for sending audio-video capture devices be converted into standard to
It identifies voice messaging, and sends it to speech recognition server；
Speech recognition server, for training acoustic model, and voice messaging and the acoustic model phase to be identified that will be received
With obtaining meaning of one's words text, and meaning of one's words text is sent to voice server；
Voice server, the meaning of one's words text for will receive are and video information and audio according to timeline information classification storage
Information distinguishes corresponding keyword, and obtained keyword is sent to keyword database；
Information server, for extracting corresponding key in keyword data according to the current time axis information received
Word, and goods matching is carried out in merchandising database according to the keyword extracted and recommends to obtain corresponding Recommendations information,
It is then sent to commercial product recommending terminal and carries out display output；
And commercial product recommending terminal, the Recommendations information for will receive carry out display output.
Preferably, the audio-video capture devices, the information server and the commercial product recommending terminal are sequentially connected electrically.
Preferably, the preprocessing server, the speech recognition server and the voice server are sequentially connected electrically,
The preprocessing server is electrically connected with the audio-video capture devices.
Preferably, the smart shopper system based on dark data mining further include: the quotient for store commodity information
Product database.
Preferably, the meaning of one's words server is electrically connected with the keyword database, the keyword database and described
Merchandising database is electrically connected with the information server respectively.
Preferably, the tonic train resolved into the voice messaging to be identified of standard including video data, and the audio
Comprising the timeline information with audio sync in sequence.
It is that the present invention generates the utility model has the advantages that
Smart shopper system and method disclosed in this invention based on dark data mining, comprising: audio-video capture devices are used for
The current time axis information in currently playing audio/video information is obtained, and sends it to information server and is handled；Letter
Server is ceased, for extracting corresponding keyword, and root in keyword data according to the current time axis information received
Goods matching is carried out in merchandising database according to the keyword extracted to recommend to obtain corresponding Recommendations information, is then sent
Display output is carried out to commercial product recommending terminal；And commercial product recommending terminal, the Recommendations information for will receive are shown
Show output；Video is changed into tonic train using NLP, then is parsed into the corresponding meaning of one's words keyword sequences of time shaft, makes it
It, can by corresponding to meaning of one's words keyword by current time axis, carrying out the matched commercial product recommending of corresponding height in video playing
The covering of high efficiency height using in video in the market, by its dark data " lighting ", realizes the intelligence of shopping information at low cost
It promotes.
Detailed description of the invention
Fig. 1, which is that the structure of the smart shopper system of the invention based on dark data mining is mad, to be pushed away；
Fig. 2 is the flow diagram of the smart shopper method of the invention based on dark data mining.
Specific embodiment
Technical solution of the present invention is clearly and completely described below in conjunction with attached drawing, it is clear that described implementation
Example is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill
Personnel's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that term " center ", "upper", "lower", "left", "right", "vertical",
The orientation or positional relationship of the instructions such as "horizontal", "inner", "outside" be based on the orientation or positional relationship shown in the drawings, merely to
Convenient for description the present invention and simplify description, rather than the device or element of indication or suggestion meaning must have a particular orientation,
It is constructed and operated in a specific orientation, therefore is not considered as limiting the invention.In addition, term " first ", " second ",
" third " is used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance.
In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can
To be mechanical connection, it is also possible to be electrically connected；It can be directly connected, can also can be indirectly connected through an intermediary
Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
Embodiment one
It is shown in Figure 1, a kind of smart shopper system based on dark data mining, comprising: audio-video capture devices, for obtaining
Current time axis information in currently playing audio/video information, and send it to information server and handled；Information clothes
It is engaged in device, for according to the current time axis information that receives, extracting corresponding keyword in keyword data, and according to mentioning
The keyword got carries out goods matching in merchandising database to be recommended to obtain corresponding Recommendations information, is then sent to quotient
Product recommend terminal to carry out display output；And commercial product recommending terminal, the Recommendations information for will receive show defeated
Out.The audio-video capture devices, the information server and the commercial product recommending terminal are sequentially connected electrically.
It is shown in Figure 1, the smart shopper system based on dark data mining further include: preprocessing server is used
Be converted into the voice messaging to be identified of standard in the video information and audio-frequency information for sending audio-video capture devices, and by its
It is sent to speech recognition server；Speech recognition server is believed for training acoustic model, and by the voice to be identified received
Breath and acoustic model match to obtain meaning of one's words text, and meaning of one's words text is sent to voice server；Voice server, being used for will
Temporally axis information classification storage is keyword corresponding with video information and audio-frequency information difference to the meaning of one's words text received.
Include the tonic train that resolves into of video data in the voice messaging to be identified of standard, and includes and sound in the tonic train
The synchronous timeline information of frequency.The preprocessing server, the speech recognition server and the voice server are successively electric
Connection, the preprocessing server are electrically connected with the audio-video capture devices.The intelligence purchase based on dark data mining
Object system further include: distinguish the keyword database of corresponding keyword for storing with video information and audio-frequency information and be used for
The merchandising database of store commodity information.The meaning of one's words server is electrically connected with the keyword database, the keyword number
It is electrically connected respectively with the information server according to library and the merchandising database.
Based on the smart shopper system of dark data mining described in the present embodiment, comprising: audio-video capture devices are used for
The current time axis information in currently playing audio/video information is obtained, and sends it to information server and is handled；Letter
Server is ceased, for extracting corresponding keyword, and root in keyword data according to the current time axis information received
Goods matching is carried out in merchandising database according to the keyword extracted to recommend to obtain corresponding Recommendations information, is then sent
Display output is carried out to commercial product recommending terminal；And commercial product recommending terminal, the Recommendations information for will receive are shown
Show output；Video is changed into tonic train using NLP, then is parsed into the corresponding meaning of one's words keyword sequences of time shaft, makes it
It, can by corresponding to meaning of one's words keyword by current time axis, carrying out the matched commercial product recommending of corresponding height in video playing
The covering of high efficiency height using in video in the market, by its dark data " lighting ", realizes the intelligence of shopping information at low cost
It promotes.
The course of work of smart shopper system described in the present embodiment based on dark data mining are as follows: preprocessing server
It converts video and audio to the voice to be identified of standard, any video data is resolved into tonic train；The tonic train
In include audio sync timeline information；Training corpus is trained to acoustic model using NLP by speech recognition server, and will
Voice and acoustics Model Matching to be identified obtains meaning of one's words text；Meaning of one's words server by meaning of one's words text time axis classification storage video/
The corresponding keyword of audio；When video playing is to corresponding time shaft, searched in information server obtained above current
Time shaft meaning of one's words keyword carries out goods matching recommendation by different dimensions (place, commodity, brand, food), and is shown
Output.
Embodiment two, the present embodiment are improved on the basis of example 1, contents described in embodiment one
It is no longer specifically to be repeated herein possessed by the present embodiment.
It is shown in Figure 2, it is a kind of realize described in the smart shopper system based on dark data mining smart shopper method,
The following steps are included:
The current time axis information in currently playing audio/video information is obtained using audio-video capture devices, and is sent it to
Information server is handled；
Using information server according to the current time axis information received, corresponding key is extracted in keyword data
Word, and goods matching is carried out in merchandising database according to the keyword extracted and recommends to obtain corresponding Recommendations information,
It is then sent to commercial product recommending terminal and carries out display output；
The Recommendations information received is subjected to display output using commercial product recommending terminal.
The smart shopper method based on dark data mining further include:
Using preprocessing server by video information and audio-frequency information that audio-video capture devices are sent be converted into standard to
It identifies voice messaging, and sends it to speech recognition server；
Using speech recognition server training acoustic model, and the voice messaging and acoustic model to be identified that receive are matched
Meaning of one's words text is obtained, and meaning of one's words text is sent to voice server；
Using voice server, by the meaning of one's words text received, temporally axis information classification storage is to believe with video information and audio
The corresponding keyword of breath difference.
The smart shopper method based on dark data mining further include:
Keyword corresponding with video information and audio-frequency information difference is stored using keyword database,
Utilize merchandising database store commodity information.
Based on the smart shopper method of dark data mining described in the present embodiment, comprising: audio-video capture devices are used for
The current time axis information in currently playing audio/video information is obtained, and sends it to information server and is handled；Letter
Server is ceased, for extracting corresponding keyword, and root in keyword data according to the current time axis information received
Goods matching is carried out in merchandising database according to the keyword extracted to recommend to obtain corresponding Recommendations information, is then sent
Display output is carried out to commercial product recommending terminal；And commercial product recommending terminal, the Recommendations information for will receive are shown
Show output；Video is changed into tonic train using NLP, then is parsed into the corresponding meaning of one's words keyword sequences of time shaft, makes it
It, can by corresponding to meaning of one's words keyword by current time axis, carrying out the matched commercial product recommending of corresponding height in video playing
The covering of high efficiency height using in video in the market, by its dark data " lighting ", realizes the intelligence of shopping information at low cost
It promotes.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (8)
1. a kind of smart shopper method based on dark data mining, it is characterised in that: the following steps are included:
Obtain video information and audio-frequency information；
The video information and audio-frequency information that will acquire are converted into the voice messaging to be identified of standard；
Voice messaging to be identified and acoustic model are matched to obtain meaning of one's words text；
By the meaning of one's words text received, temporally axis information classification storage is corresponding with video information and audio-frequency information difference
Keyword；
Keyword is constituted into keyword database；
According to the timeline information of current video information and audio-frequency information, corresponding keyword is extracted in keyword data；
Goods matching is carried out in merchandising database according to the keyword extracted to recommend to obtain corresponding Recommendations information simultaneously
Carry out display output.
2. the smart shopper method according to claim 1 based on dark data mining, it is characterised in that: standard it is to be identified
It include the tonic train that video data resolves into voice messaging, and comprising the time with audio sync in the tonic train
Axis information.
3. a kind of smart shopper system for realizing the smart shopper method based on dark data mining as described in claim 1, special
Sign is: including:
Audio-video capture devices, for obtaining video information and audio-frequency information, and the video information and audio-frequency information that will acquire are pre-
Processing server；Meanwhile the current time axis information in currently playing audio/video information is obtained, and send it to information clothes
Business device is handled；
Preprocessing server, video information and audio-frequency information for sending audio-video capture devices be converted into standard to
It identifies voice messaging, and sends it to speech recognition server；
Speech recognition server, for training acoustic model, and voice messaging and the acoustic model phase to be identified that will be received
With obtaining meaning of one's words text, and meaning of one's words text is sent to voice server；
Voice server, the meaning of one's words text for will receive are and video information and audio according to timeline information classification storage
Information distinguishes corresponding keyword, and obtained keyword is sent to keyword database；
Information server, for extracting corresponding key in keyword data according to the current time axis information received
Word, and goods matching is carried out in merchandising database according to the keyword extracted and recommends to obtain corresponding Recommendations information,
It is then sent to commercial product recommending terminal and carries out display output；
And commercial product recommending terminal, the Recommendations information for will receive carry out display output.
4. the smart shopper system according to claim 3 based on dark data mining, it is characterised in that: the audio-video obtains
Device, the information server and the commercial product recommending terminal is taken to be sequentially connected electrically.
5. the smart shopper system according to claim 4 based on dark data mining, it is characterised in that: the pretreatment clothes
Business device, the speech recognition server and the voice server are sequentially connected electrically, and the preprocessing server and the sound regard
The electrical connection of frequency acquisition device.
6. the smart shopper system according to claim 5 based on dark data mining, it is characterised in that: further include for depositing
Store up the merchandising database of merchandise news.
7. the smart shopper system according to claim 6 based on dark data mining, it is characterised in that: the meaning of one's words service
Device is electrically connected with the keyword database, the keyword database and the merchandising database respectively with the information service
Device electrical connection.
8. the smart shopper system according to claim 7 based on dark data mining, it is characterised in that: standard it is to be identified
It include the tonic train that video data resolves into voice messaging, and comprising the time with audio sync in the tonic train
Axis information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811598887.3A CN109785052A (en) | 2018-12-26 | 2018-12-26 | Smart shopper method and system based on dark data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811598887.3A CN109785052A (en) | 2018-12-26 | 2018-12-26 | Smart shopper method and system based on dark data mining |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109785052A true CN109785052A (en) | 2019-05-21 |
Family
ID=66498387
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811598887.3A Pending CN109785052A (en) | 2018-12-26 | 2018-12-26 | Smart shopper method and system based on dark data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109785052A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340550A (en) * | 2020-02-27 | 2020-06-26 | 谷元(上海)文化科技有限责任公司 | Data information extraction method |
| CN113891119A (en) * | 2021-09-28 | 2022-01-04 | 北京字跳网络技术有限公司 | Information display method and device, electronic equipment and storage medium |
| CN115134636A (en) * | 2022-08-29 | 2022-09-30 | 北京达佳互联信息技术有限公司 | Information recommendation method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281534A (en) * | 2008-05-28 | 2008-10-08 | 叶睿智 | Method for searching multimedia resource based on audio content retrieval |
| CN105657563A (en) * | 2014-11-12 | 2016-06-08 | 深圳富泰宏精密工业有限公司 | Commodity recommending system and method based on video contents |
| US20180108048A1 (en) * | 2016-10-19 | 2018-04-19 | Samsung Sds Co., Ltd. | Method, apparatus and system for recommending contents |
| 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 |
-
2018
- 2018-12-26 CN CN201811598887.3A patent/CN109785052A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281534A (en) * | 2008-05-28 | 2008-10-08 | 叶睿智 | Method for searching multimedia resource based on audio content retrieval |
| CN105657563A (en) * | 2014-11-12 | 2016-06-08 | 深圳富泰宏精密工业有限公司 | Commodity recommending system and method based on video contents |
| US20180108048A1 (en) * | 2016-10-19 | 2018-04-19 | Samsung Sds Co., Ltd. | Method, apparatus and system for recommending contents |
| 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 |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340550A (en) * | 2020-02-27 | 2020-06-26 | 谷元(上海)文化科技有限责任公司 | Data information extraction method |
| CN113891119A (en) * | 2021-09-28 | 2022-01-04 | 北京字跳网络技术有限公司 | Information display method and device, electronic equipment and storage medium |
| WO2023051297A1 (en) * | 2021-09-28 | 2023-04-06 | 北京字跳网络技术有限公司 | Information display method and apparatus, electronic device, and storage medium |
| CN115134636A (en) * | 2022-08-29 | 2022-09-30 | 北京达佳互联信息技术有限公司 | Information recommendation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Videoclip: Contrastive pre-training for zero-shot video-text understanding | |
| CN112015949A (en) | Video generation method and device, storage medium and electronic equipment | |
| CN107862027A (en) | Retrieve intension recognizing method, device, electronic equipment and readable storage medium storing program for executing | |
| US20100223335A1 (en) | Dynamically Managing Online Communication Groups | |
| KR102576344B1 (en) | Method and apparatus for processing video, electronic device, medium and computer program | |
| CN108388650B (en) | Search processing method and device based on requirements and intelligent equipment | |
| CN108416041B (en) | Voice log analysis method and system | |
| CN109785052A (en) | Smart shopper method and system based on dark data mining | |
| CN105095288A (en) | Data analysis method and data analysis device | |
| Ilina et al. | Social event detection on twitter | |
| CN112738556A (en) | Video processing method and device | |
| KR102606175B1 (en) | Method and device for pushing information | |
| CN104142936A (en) | Audio and video match method and audio and video match device | |
| CN113806588A (en) | Method and device for searching video | |
| CN112541125A (en) | Sequence labeling model training method and device and electronic equipment | |
| CN105045882B (en) | A kind of hot word processing method and processing device | |
| CN108846098A (en) | A kind of information flow summarization generation and methods of exhibiting | |
| US20240220537A1 (en) | Metadata tag identification | |
| CN115098729A (en) | Video processing method, sample generation method, model training method and device | |
| CN108108391A (en) | For the processing method and device of the information of data visualization | |
| CN108090201A (en) | A kind of method, apparatus and electronic equipment of article content classification | |
| JP6341899B2 (en) | Advertisement system and advertisement delivery method | |
| EP3905060A1 (en) | Artificial intelligence for content discovery | |
| CN112446214A (en) | Method, device and equipment for generating advertisement keywords and storage medium | |
| CN111353070A (en) | Video title processing method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190521 |