CN115086721B - Ultra-high-definition live system service supervision system based on data analysis - Google Patents
Ultra-high-definition live system service supervision system based on data analysis Download PDFInfo
- Publication number
- CN115086721B CN115086721B CN202211002379.0A CN202211002379A CN115086721B CN 115086721 B CN115086721 B CN 115086721B CN 202211002379 A CN202211002379 A CN 202211002379A CN 115086721 B CN115086721 B CN 115086721B
- Authority
- CN
- China
- Prior art keywords
- selling
- commodity
- target
- live broadcast
- display
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- 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
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/233—Processing of audio elementary streams
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/254—Management at additional data server, e.g. shopping server, rights management server
- H04N21/2542—Management at additional data server, e.g. shopping server, rights management server for selling goods, e.g. TV shopping
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
- H04N21/4788—Supplemental services, e.g. displaying phone caller identification, shopping application communicating with other users, e.g. chatting
Abstract
The invention relates to the technical field of live broadcast supervision, and particularly discloses a data analysis-based ultra-high-definition live broadcast system service supervision system, which adopts an online intelligent supervision mode when carrying out illegal sale supervision on ultra-high-definition live broadcast tape goods videos, extracts live broadcast sale information corresponding to currently sold goods from the live broadcast ultra-high-definition live broadcast tape goods videos, evaluates a live broadcast sale compliance index corresponding to the currently sold goods, carries out targeted processing on a live broadcast process of a specified ultra-high-definition live broadcast room by a live broadcast platform, realizes effective supervision on the live broadcast tape goods videos, reports the supervision for a user, can directly extract illegal sale evidences, improves the supervision timeliness, reduces the collection cost of illegal evidences, reduces the investigation and evidence taking difficulty on the one hand, avoids the occurrence of invalid supervision phenomenon on the other hand, and greatly meets the current sale supervision requirements of the illegal sale of the ultra-high-definition live broadcast tape goods videos, convenience and cost saving on the other hand.
Description
Technical Field
The invention relates to the technical field of live broadcast supervision, in particular to an ultra-high-definition live broadcast system service supervision system based on data analysis.
Background
In a new era with the daily change of internet technology, the form of live webcasting is created, and meanwhile, under the background of the new era, the development of ultra-high definition video industry is promoted by the increasing maturity of the 5G technology, so that the live webcasting is widely applied to the construction of ultra-high definition live webcasting systems, and the watching experience of live webcasting is better. However, with the rise and rapid development of internet technologies in recent years, internet groups are greatly expanded, more and more sellers start to transfer commodity sales places to network markets, live webcast goods becomes a main channel of commodity sales, the distance between products and users is shortened by using the interactivity and timeliness of live webcasting, the users can experience themselves in stores on line, and the shopping trust sense of the users is greatly increased.
In order to attract users and improve sales volume, the behaviors of false propaganda and counterfeit sale of sold commodities often exist in the ultra-high definition live broadcast room at present when live broadcast strip goods are carried, the benefits of consumers are greatly damaged, and in view of the situation, illegal sale supervision of ultra-high definition live broadcast strip goods videos is very important.
However, at present, a user reporting mode is basically adopted for the illegal selling supervision mode of the ultra-high-definition live video with goods, and the supervision mode needs to spend time for checking and evidence obtaining when receiving the report of the user, so that the supervision timeliness is reduced, and the evidence collection cost is increased; on the other hand, some users do not report in time when watching the live broadcast cargo carrying video of the false propaganda, but report after a period of time, report delay is caused to a certain extent, loss rate of the users is increased, the possibility that the live broadcast cargo carrying video reported in the process is deleted exists, great difficulty is brought to investigation and evidence obtaining, on the other hand, due to the fact that authenticity of report of the users is not studied, invalid supervision is easily caused, and supervision resources are wasted.
In conclusion, at present, illegal selling supervision of ultrahigh-definition live broadcast video with goods is carried out according to a user reporting mode, the live broadcast video with goods cannot be effectively supervised, and the current illegal selling supervision requirements for timely, convenient and cost-saving ultrahigh-definition live broadcast video with goods cannot be met.
Disclosure of Invention
In order to solve the technical problems, the invention is realized by the following technical scheme: an ultra high definition live system service supervision system based on data analysis comprises: and the commodity selling list extracting module is used for recording the ultra high definition live broadcast room to be supervised as the appointed ultra high definition live broadcast room, further extracting commodity selling lists corresponding to the current live broadcast of the appointed ultra high definition live broadcast room from the live broadcast platform background, and extracting each selling commodity name and the selling starting time point from the commodity selling lists.
The live broadcast video monitoring module is used for acquiring current time points in real time in a live broadcast process corresponding to the appointed ultra-high-definition live broadcast room, comparing the current time points with the starting selling time points corresponding to the sold commodities, recording the current live broadcast video as the live broadcast video when the starting selling time points corresponding to the sold commodities are reached, recording the sold commodities as target sold commodities, and monitoring the live broadcast video.
And the target selling commodity display information extraction module is used for extracting display information corresponding to the target selling commodity from the key live broadcast video.
And the target selling commodity description audio information extraction module is used for extracting the description audio information of the anchor on the target selling commodity from the key live broadcast video.
And the supervision service library is used for storing the importance degrees corresponding to various commodity attributes, storing live broadcast selling compliance index intervals corresponding to various live broadcast selling compliance grades, and storing live broadcast process processing modes corresponding to various live broadcast selling compliance grades.
And the target selling commodity public screen question reply information extraction module is used for extracting reply information of the main broadcast to the question information from the public screen of the key live broadcast video.
The target selling commodity live selling conformity analysis module is used for respectively carrying out real object display conformity, description attribute conformity, description language conformity and anchor reply conformity analysis on display information corresponding to the target selling commodity, description audio information of the anchor selling commodity and reply information of the anchor to question information.
The target selling commodity direct broadcast selling compliance comprehensive evaluation module is used for comprehensively evaluating a direct broadcast selling compliance index corresponding to the target selling commodity.
And the direct broadcasting process processing module of the appointed ultra-high-definition direct broadcasting room is used for carrying out targeted processing on the direct broadcasting process of the appointed ultra-high-definition direct broadcasting room by the direct broadcasting platform according to the direct broadcasting selling compliance index corresponding to the target selling commodity.
Further optimizing the technical scheme, the display information comprises a real object display image and a shopping display link.
Further optimize this technical scheme, the commodity is sold in the target and the direct broadcast of selling accords with degree analysis module includes that the commodity show accord with degree analysis unit is sold in the target, the commodity is sold in the target and the description audio frequency accords with degree analysis unit is sold in the target and the commodity reply accords with degree analysis unit is sold in the target.
Further optimize this technical scheme, the target is sold commodity show conformity analysis unit and is used for selling the show image in kind that the commodity corresponds to the target and carrying out show conformity analysis in kind, and its concrete analytic process is as follows: and clicking a shopping display link corresponding to the target selling commodity, and further extracting detail page appearance display information of the target selling commodity in the shopping display link, wherein the detail page appearance display information comprises page display size specifications and page display color chromaticity.
And carrying out page display outline sketching according to the page display size specification corresponding to the target selling commodity, and acquiring a page display outline surface area corresponding to the target selling commodity based on the page display outline sketching.
And extracting the physical outline from the physical display image of the target sold commodity, acquiring the camera shooting proportion of the physical display image, and then carrying out scaling on the physical outline corresponding to the target sold commodity according to the camera shooting proportion to obtain a physical reference outline corresponding to the target sold commodity.
And carrying out coincidence comparison on the real object reference outline corresponding to the target sold commodity and the page display outline to obtain the coincidence area of the real object outlines.
Displaying the coincidence area of the outline of the object corresponding to the target commodity and the surface area of the display outline of the page through the appearance of the objectEquation for calculating apparent coefficient
And obtaining the appearance display fit coefficient of the object corresponding to the target sold commodity.
Extracting the color chroma of the object from the object display image of the object commodity, comparing the color chroma with the page display color chroma corresponding to the object commodity, and displaying the coincidence coefficient calculation formula according to the color of the object
And obtaining a color display coincidence coefficient of a real object corresponding to the target selling commodity.
Calculating the object display conformity degree corresponding to the target commodity according to the object appearance display conformity coefficient and the object color display conformity coefficient corresponding to the target commodity
The calculation formula is
Wherein, A and B are respectively expressed as the proportion factor corresponding to the real object appearance display coincidence coefficient and the real object color display coincidence coefficient, and e is expressed as a natural constant.
Further optimizing the technical solution, the target selling commodity description audio conformity analyzing unit is used for performing description attribute conformity analysis on the description audio information of the target selling commodity by the anchor, and the specific analysis process is as follows: and carrying out voice recognition on the description audio information of the target selling commodity by the anchor to obtain the description text information of the target selling commodity by the anchor.
And extracting description attribute information from the description text information of the anchor on the target selling goods, and counting the number of the description attributes and specific information corresponding to various description attributes.
And counting the number of the display attributes and the specific information corresponding to various display attributes from the attribute display information of the detail page of the target selling commodity in the shopping display link.
Matching various description attributes corresponding to the target selling goods with various display attributes, counting the number of the description attributes successfully matched, and recording the description attributes successfully matched as matching description attributes.
Comparing the specific information of each matching description attribute with the specific information of the corresponding display attribute, and recording the description coincidence coefficient corresponding to a certain matching description attribute as the specific information of the corresponding display attribute if the specific information of the matching description attribute is consistent with the specific information of the corresponding display attribute
If not, the description coincidence coefficient corresponding to the matching description attribute is recorded as
。
And extracting the importance corresponding to the various matching description attributes from the supervision service library based on the various matching description attributes corresponding to the target selling goods.
Counting the corresponding description attribute conformity of the target selling goods according to the number of the matching description attributes and the description conformity coefficient and the importance degree corresponding to each matching description attribute
The calculation formula is
K is expressed as the number of match description attributes, K is expressed as the number of presentation attributes,
the description corresponding to the matching description attribute represented as the ith is consistent with the coefficient, i is represented as the number of the matching description attribute, i =1,2
Is taken as
Or
，
And expressing the importance corresponding to the ith matching description attribute.
Further optimizing the technical solution, the target selling goods description audio conformity analysis unit is further configured to perform description term conformity analysis on description audio information of the target selling goods by the anchor, and the specific analysis process is as follows: and the anchor carries out stop word and word segmentation processing on the description text information of the target selling goods to obtain a plurality of description word groups.
And matching each description phrase with the constructed irregular publicity phrase vocabulary library, counting the number of description phrases successfully matched according to the matching, and marking the number as the number of illegal description phrases.
Calculating the conformity of descriptive phrases corresponding to the target selling goods based on the number of descriptive phrases corresponding to the target selling goods and the number of illegal descriptive phrases
The calculation formula is
。
Further optimizing the technical solution, the objective selling commodity reply conformity analysis unit is used for performing anchor reply conformity analysis on reply information of the challenge information by the anchor, and the specific analysis method is as follows: and counting and watching the number of question attributes of the target selling goods from the question information.
And extracting specific reply information of the anchor to various question attributes from public screen reply information of the anchor to the question information.
And extracting specific display information corresponding to various question attributes from the detail page attribute display information corresponding to the target selling commodity according to various question attributes of the watching user to the target selling commodity.
And comparing the specific reply information of the anchor to various question attributes with the specific display information corresponding to the various question attributes, counting the number of the question attributes which are consistent in comparison, and recording the question attributes which are consistent in comparison as the adaptive question attributes.
And matching the adaptive question attributes with the importance degrees corresponding to the attributes of various commodities in the supervision service library, and matching the importance degrees corresponding to the adaptive question attributes.
Calculating the anchor reply conformity degree corresponding to the target selling commodity based on the question attribute quantity, the adaptive question attribute quantity and the importance degree corresponding to each adaptive question attribute of the watching user to the target selling commodity
Wherein
X represents the number of questioning attributes of the viewing user for the target offered merchandise,
the importance degree corresponding to the jth type of the adapted question attribute is represented, j is represented as the number of the adapted question attribute, and j =1, 2.
Further optimizing the technical scheme, the direct broadcast selling compliance index evaluation formula corresponding to the target selling commodity is
，
The direct broadcast selling compliance index corresponding to the target selling commodity is expressed, and a, b, c and d respectively express weight coefficients corresponding to the set real object display conformity, description attribute conformity, description phrase conformity and anchor reply conformity.
Further optimizing the technical scheme, the specific processing steps corresponding to the targeted processing of the live broadcast process of the designated ultra-high-definition live broadcast room are as follows: sell the direct broadcast that commodity corresponds and sell the compliance index and the supervision service library in various direct broadcasts sell the compliance index interval that the compliance grade corresponds and contrast, select the direct broadcast that the commodity corresponds and sell the compliance grade from it to sell the target.
And matching the live broadcast selling compliance grade corresponding to the target selling commodity with the live broadcast process processing modes corresponding to various live broadcast selling compliance grades in the supervision service library, so that the live broadcast process processing modes corresponding to the appointed ultra-high-definition live broadcast room are matched and executed by a live broadcast platform.
Further optimize this technical scheme, the live broadcast process treatment mode that the compliance level corresponds is sold in various live broadcasts includes continuation normal live broadcast, interruption prompt live broadcast and live broadcast permission forbidding.
Compared with the prior art, the invention has the following advantages: 1. according to the method, an online intelligent supervision mode is adopted when illegal selling supervision is carried out on the ultrahigh-definition live broadcast video, live broadcast selling information corresponding to the current sold goods is extracted from the live broadcast video, and therefore the live broadcast selling compliance index corresponding to the current sold goods is evaluated, so that a live broadcast process of a specified ultrahigh-definition live broadcast room is processed in a targeted mode through a live broadcast platform, effective supervision on the live broadcast video is achieved, and compared with the supervision reported by users, the supervision mode can directly extract illegal selling evidences, not only is the supervision timeliness improved, but also the collection cost of illegal selling evidences is reduced, meanwhile, on one hand, investigation evidence obtaining difficulty is reduced, on the other hand, the selling evidences can be extracted from the live broadcast video, authenticity of the extracted evidences is guaranteed to the greatest extent, waste of supervision resources is reduced, the occurrence of invalid supervision phenomena is avoided, and illegal selling requirements for timely, convenient and cost-saving of the ultrahigh-definition live broadcast video are greatly met.
2. According to the invention, when the direct broadcast selling compliance index corresponding to the current sold commodity is evaluated, the direct broadcast selling compliance index corresponding to the current sold commodity is evaluated respectively from multiple dimensions of the real object display compliance, the description attribute compliance, the description term compliance and the anchor reply compliance, so that the comprehensive evaluation of the sold commodity compliance is realized, the evaluation content range is expanded by increasing the evaluation dimension, the evaluation deviation caused by single evaluation dimension is avoided on one hand, and the omission evaluation that some illegal sold commodities exist in certain aspects and do not exist in other aspects is avoided on the other hand, therefore, the accurate reliability of the evaluation result is improved, the continuous selling rate of the illegal sold commodities is reduced, the supervision coverage range of the direct broadcast commodities is expanded to a certain extent, the direct broadcast selling compliance index is beneficial to the fast and whole illegal selling market, and the direct broadcast selling compliance index corresponding to the current sold commodity is higher in practicability.
Drawings
The invention is further illustrated by means of the attached drawings, but the embodiments in the drawings do not constitute any limitation to the invention, and for a person skilled in the art, without inventive effort, further drawings may be derived from the following figures.
Fig. 1 is a schematic diagram of system module connection according to the present invention.
Fig. 2 is a schematic view of the connection of the target selling goods direct broadcast selling conformity analysis module of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Referring to fig. 1, a super high definition live broadcast system service supervision system based on data analysis comprises a commodity selling list extraction module of a designated super high definition live broadcast room, an important live broadcast video monitoring module, a target selling commodity display information extraction module, a target selling commodity description audio information extraction module, a supervision service library, a target selling commodity public screen question reply information extraction module, a target selling commodity live broadcast selling conformity degree analysis module, a target selling commodity live broadcast selling conformity comprehensive evaluation module and a designated super high definition live broadcast room live broadcast process processing module.
Commodity selling inventory of appointed super high definition live broadcast room in the above-mentioned is got module and is connected with key live broadcast video monitoring module, key live broadcast video monitoring module is respectively sold commodity show information with the target and is drawed the module, the target is sold commodity description audio information and is drawed module and the target is sold commodity public screen and ask reply information to draw the module and be connected, the target is sold commodity show information and is drawed the module, the target is sold commodity description audio information and is drawn the module, the supervision service library and the target is sold commodity public screen and ask reply information to draw the module all to sell the degree of conformity analysis module with the target and be connected with the target and sell commodity live broadcast and sell conformity comprehensive evaluation module, the target is sold commodity and is sold live broadcast and sell conformity comprehensive evaluation module and supervision service library and all with appointed super high definition live broadcast room live broadcast process processing module and be connected.
The commodity selling list extracting module of the appointed ultra-high-definition live broadcast room is used for recording the ultra-high-definition live broadcast room to be supervised as the appointed ultra-high-definition live broadcast room, further extracting a commodity selling list corresponding to the current live broadcast of the appointed ultra-high-definition live broadcast room from a live broadcast platform background, and extracting each selling commodity name and a selling starting time point from the commodity selling list.
The live broadcast video monitoring module is used for acquiring a current time point in real time in a live broadcast process corresponding to a specified ultra-high definition live broadcast room, comparing the current time point with a starting selling time point corresponding to each sold commodity, recording the current live broadcast video as a live broadcast video when the starting selling time point corresponding to a certain sold commodity is reached, recording the sold commodity as a target sold commodity, and monitoring the live broadcast video.
The target selling commodity display information extraction module is used for extracting display information corresponding to the target selling commodity from the key live broadcast video, wherein the display information comprises a real object display image and a shopping display link.
The target selling commodity description audio information extraction module is used for extracting description audio information of the target selling commodity from the key live broadcast video.
The supervision service library is used for storing importance degrees corresponding to various commodity attributes, storing live broadcast selling compliance index intervals corresponding to various live broadcast selling compliance grades, and storing live broadcast process processing modes corresponding to various live broadcast selling compliance grades, wherein the live broadcast process processing modes corresponding to various live broadcast selling compliance grades comprise continuous normal live broadcast, interruption prompt live broadcast and live broadcast permission seal prohibition.
The target selling commodity public screen question reply information extraction module is used for extracting reply information of a main broadcast to the question information from a public screen of a key live broadcast video.
The target selling commodity live broadcast selling conformity analysis module is used for respectively carrying out real object display conformity, description attribute conformity, description phrase conformity and anchor reply conformity analysis on display information corresponding to the target selling commodity, description audio information of the anchor selling commodity and reply information of the anchor to question information.
In a preferred technical solution of the present invention, referring to fig. 2, the target sold commodity live-broadcast selling conformity analysis module includes a target sold commodity physical display conformity analysis unit, a target sold commodity description audio conformity analysis unit, and a target sold commodity reply conformity analysis unit.
The target commodity selling real object display conformity analysis unit is used for carrying out real object display conformity analysis on a real object display image corresponding to the target commodity selling, and the specific analysis process is as follows: and clicking a shopping display link corresponding to the target commodity to be sold, and further extracting detail page appearance display information of the target commodity to be sold in the shopping display link, wherein the detail page appearance display information comprises page display size specifications and page display color chromaticity.
And carrying out page display outline sketching according to the page display size specification corresponding to the target selling commodity, and acquiring a page display outline surface area corresponding to the target selling commodity based on the page display outline sketching.
And extracting the physical outline from the physical display image of the target sold commodity, acquiring the camera shooting proportion of the physical display image, and then carrying out scaling on the physical outline corresponding to the target sold commodity according to the camera shooting proportion to obtain a physical reference outline corresponding to the target sold commodity.
And carrying out coincidence comparison on the real object reference outline corresponding to the target sold commodity and the page display outline to obtain the coincidence area of the real object outlines.
Calculating the coincidence area of the outline of the object corresponding to the target commodity and the surface area of the display outline of the page through the coincidence coefficient of the display outline of the object
And obtaining a real object appearance display coincidence coefficient corresponding to the target selling commodity, wherein the larger the coincidence area of the real object appearance contour is, the larger the real object appearance display coincidence coefficient is.
Extracting object color chroma from an object display image of the object commodity, comparing the object color chroma with page display color chroma corresponding to the object commodity, and displaying a coincidence coefficient calculation formula through the object color
And obtaining a real object color display coincidence coefficient corresponding to the target commodity, wherein the difference between the real object color chromaticity and the page display color chromaticity is smaller, and the real object color display coincidence coefficient is larger.
Calculating the object display conformity degree corresponding to the target commodity according to the object appearance display conformity coefficient and the object color display conformity coefficient corresponding to the target commodity
The calculation formula is
Wherein, A and B are respectively expressed as the proportion factor corresponding to the real object appearance display coincidence coefficient and the real object color display coincidence coefficient, and e is expressed as a natural constant.
It should be noted that the reason why the object display conformity analysis is performed on the target sold product in the invention is that in order to attract the audience when some products are sold by the present anchor, the object of the product displayed to the audience is inconsistent with the product displayed in the purchase link on the upper shelf, which leads to the serious disagreement between the product ordered by the audience and the displayed product.
The target selling commodity description audio conformity analysis unit is used for carrying out description attribute conformity analysis on description audio information of a target selling commodity by an anchor, and the specific analysis process is as follows: and carrying out voice recognition on the description audio information of the target selling commodity by the anchor to obtain the description text information of the target selling commodity by the anchor.
And extracting description attribute information from the description text information of the anchor on the target selling goods, and counting the number of the description attributes and specific information corresponding to various description attributes.
In one embodiment, the aforementioned attributes include brand, size, material, price, weight, production area, composition, applicable population, etc., and the specific information corresponding to various descriptive attributes, for example, the specific information corresponding to the material attribute is cotton and hemp, and the specific information corresponding to the weight attribute is 200g.
And counting the number of the display attributes and the specific information corresponding to various display attributes from the attribute display information of the detail page of the target commodity in the shopping display link.
And matching various description attributes corresponding to the target selling commodity with various display attributes, counting the number of the description attributes successfully matched, and recording the description attributes successfully matched as matching description attributes.
Comparing the specific information of each matching description attribute with the specific information of the corresponding display attribute, and recording the description coincidence coefficient corresponding to a certain matching description attribute as the specific information of the corresponding display attribute if the specific information of the matching description attribute is consistent with the specific information of the corresponding display attribute
If not, the description coincidence coefficient corresponding to the matching description attribute is recorded as
Wherein
。
And extracting the importance corresponding to the various matching description attributes from the supervision service library based on the various matching description attributes corresponding to the target selling goods.
Counting the corresponding description attribute conformity of the target selling commodity according to the number of the matching description attributes and the corresponding description conformity coefficient and the corresponding importance degree of each matching description attribute
The calculation formula is
K is expressed as the number of matching description attributes, K is expressed as the number of presentation attributes,
the description corresponding to the matching description attribute represented as the ith is consistent with the coefficient, i is represented as the number of the matching description attribute, i =1,2
Is taken as
Or
，
Expressed as the importance corresponding to the ith matching description attributeAnd (4) degree.
The target selling commodity description audio conformity analysis unit is also used for carrying out description word conformity analysis on description audio information of the target selling commodity by the anchor, and the specific analysis process is as follows: and the anchor carries out stop word and word segmentation processing on the description text information of the target selling goods to obtain a plurality of description word groups.
And matching each description phrase with the constructed non-standard publicity phrase vocabulary library, counting the number of description phrases successfully matched according to the matching, and marking the number as the number of violation description phrases.
The above-mentioned non-compliance glossary libraries include top, best, top, national, etc.
Calculating the conformity of descriptive phrases corresponding to the target selling goods based on the number of descriptive phrases corresponding to the target selling goods and the number of illegal descriptive phrases
The calculation formula is
And the less the number of the illegal description phrases is, the greater the conformity of the description phrases is.
It should be noted that the reason why the present invention analyzes the description attribute conformity and the description term conformity of the target selling goods is that the attributes of the goods are often falsely described and exaggeratedly publicized to mislead the consumer in order to stimulate the desire of the user to buy in the process of selling the goods by the current anchor, and the description attributes and the description terms of the selling goods need to be subjected to compliance analysis in order to prevent the consumer from being misled in advance in the process of selling the goods by live broadcast and to guarantee the legitimate interest of the consumer.
The reply conformity analysis unit of the target selling commodity is used for carrying out anchor reply conformity analysis on reply information of the challenge information by an anchor, and the specific analysis method is as follows: and counting and watching the number of question attributes of the target selling goods from the question information.
And extracting specific reply information of the anchor to various question attributes from public screen reply information of the anchor to the question information.
And extracting specific display information corresponding to various question attributes from the detail page attribute display information corresponding to the target selling commodity according to various question attributes of the watching user to the target selling commodity.
And comparing the specific reply information of the anchor to the various question attributes with the specific display information corresponding to the various question attributes, counting the number of the question attributes which are consistent in comparison, and marking the question attributes which are consistent in comparison as the adaptive question attributes.
And matching the adaptive question attributes with the importance degrees corresponding to the attributes of various commodities in the supervision service library, and matching the importance degrees corresponding to the adaptive question attributes.
Calculating the anchor reply conformity degree corresponding to the target selling commodity based on the question attribute quantity, the adaptive question attribute quantity and the importance degree corresponding to each adaptive question attribute of the watching user to the target selling commodity
Wherein
X represents the number of questioning attributes of the viewing user for the targeted merchandise item to be sold,
the importance degree corresponding to the jth type of adaptation question attribute is represented, j represents the number of the adaptation question attribute, j =1, 2.
It should be noted that the reason why the anchor reply conformity analysis is performed on the target sold goods in the invention is that the interaction between the user and the anchor in the current live broadcast room is frequent, so that the user can ask questions about the attribute information of the target sold goods on the live broadcast public screen frequently, and when the anchor replies the question information, in order to improve the viewing viscosity of the user to the live broadcast room, the anchor replies a false condition, so that the conformity analysis must be performed on the reply information of the anchor.
The target sold commodity direct broadcast selling compliance comprehensive evaluation module is used for comprehensively evaluating direct broadcast selling compliance indexes corresponding to the target sold commodities, and the evaluation formula is
，
The direct broadcast selling compliance index corresponding to the target selling commodity is expressed, and a, b, c and d respectively express weight coefficients corresponding to the set real object display conformity, description attribute conformity, description phrase conformity and anchor reply conformity.
According to the embodiment of the invention, when the direct broadcast selling compliance index corresponding to the current selling commodity is evaluated, the direct broadcast selling compliance index corresponding to the current selling commodity is evaluated respectively from multiple dimensions of the real object display compliance, the description attribute compliance, the description term compliance and the anchor reply compliance, so that the comprehensive evaluation of the selling commodity compliance is realized, the evaluation content range is enlarged by increasing the evaluation dimension, the evaluation deviation caused by a single evaluation dimension is avoided, and the omission evaluation that some illegal selling commodities exist in a certain aspect and do not exist in other aspects is avoided, so that the accurate reliability of the evaluation result is improved, the continuous incidence rate of the illegal selling commodities is reduced, the supervision coverage range of the direct broadcast selling commodities is enlarged to a certain extent, the direct broadcast selling compliance index is beneficial to the rapid and normalized selling market selling, and the direct broadcast selling compliance index corresponding to the current selling commodity is high in practicability.
The direct broadcasting process processing module of the appointed ultra-high-definition direct broadcasting room is used for carrying out targeted processing on a direct broadcasting process of the appointed ultra-high-definition direct broadcasting room by a direct broadcasting platform according to a direct broadcasting selling compliance index corresponding to a target selling commodity, and the specific processing steps are as follows: and the live broadcast selling compliance index corresponding to the target selling commodity is compared with the live broadcast selling compliance index interval corresponding to various live broadcast selling compliance grades in the supervision service library, and the live broadcast selling compliance grade corresponding to the target selling commodity is selected from the live broadcast selling compliance index interval.
And matching the live broadcast selling compliance grade corresponding to the target selling commodity with the live broadcast process processing modes corresponding to various live broadcast selling compliance grades in the supervision service library, so that the live broadcast process processing modes corresponding to the appointed ultra-high-definition live broadcast room are matched and executed by a live broadcast platform.
Meanwhile, when the direct broadcast process processing mode of the appointed ultra-high definition direct broadcast room is normal direct broadcast continuously, other commodities continuously sold in the appointed ultra-high definition direct broadcast room are continuously supervised according to the supervision mode until the last commodity is sold.
According to the method, an online intelligent supervision mode is adopted when illegal selling supervision is carried out on the ultrahigh-definition live broadcast video, live broadcast selling information corresponding to the current sold goods is extracted from the live broadcast video, and therefore the live broadcast selling compliance index corresponding to the current sold goods is evaluated, so that a live broadcast process of a specified ultrahigh-definition live broadcast room is processed in a targeted mode through a live broadcast platform, effective supervision on the live broadcast video is achieved, and compared with the supervision reported by users, the supervision mode can directly extract illegal selling evidences, not only is the supervision timeliness improved, but also the collection cost of illegal selling evidences is reduced, meanwhile, on one hand, investigation evidence obtaining difficulty is reduced, on the other hand, the selling evidences can be extracted from the live broadcast video, authenticity of the extracted evidences is guaranteed to the greatest extent, waste of supervision resources is reduced, the occurrence of invalid supervision phenomena is avoided, and illegal selling requirements for timely, convenient and cost-saving of the ultrahigh-definition live broadcast video are greatly met.
The foregoing is merely illustrative and explanatory of the present invention and various modifications, additions or substitutions may be made to the specific embodiments described by those skilled in the art without departing from the scope of the invention as defined in the accompanying claims.
Claims (3)
1. A super high definition live system service supervision system based on data analysis is characterized by comprising:
the system comprises a specified ultra-high-definition live broadcast room commodity selling list extracting module, a monitoring module and a monitoring module, wherein the specified ultra-high-definition live broadcast room commodity selling list extracting module is used for recording an ultra-high-definition live broadcast room to be supervised as a specified ultra-high-definition live broadcast room, further extracting a commodity selling list corresponding to the current live broadcast of the specified ultra-high-definition live broadcast room from a live broadcast platform background, and extracting each selling commodity name and a selling starting time point from the commodity selling list;
the system comprises a key live broadcast video monitoring module, a super high definition live broadcast room and a plurality of selling commodities, wherein the key live broadcast video monitoring module is used for acquiring a current time point in real time in a live broadcast process corresponding to the appointed super high definition live broadcast room, comparing the current time point with a selling starting time point corresponding to each selling commodity, recording the current live broadcast video as a key live broadcast video when the selling starting time point corresponding to a certain selling commodity is reached, and recording the selling commodity as a target selling commodity, so that the key live broadcast video is monitored;
the target selling commodity display information extraction module is used for extracting display information corresponding to the target selling commodity from the key live broadcast video;
the target selling commodity description audio information extraction module is used for extracting description audio information of the anchor on the target selling commodity from the key live broadcast video;
the monitoring service library is used for storing importance degrees corresponding to various commodity attributes, storing live broadcast selling compliance index intervals corresponding to various live broadcast selling compliance grades and storing live broadcast process processing modes corresponding to various live broadcast selling compliance grades;
the target selling commodity public screen question reply information extraction module is used for extracting reply information of the main broadcast to the question information from a public screen of the key live broadcast video;
the target selling commodity direct broadcast selling conformity analysis module is used for respectively carrying out real object display conformity, description attribute conformity, description phrase conformity and anchor reply conformity analysis on display information corresponding to the target selling commodity, description audio information of the anchor on the target selling commodity and reply information of the anchor on the question information;
the target sold commodity direct broadcast selling compliance comprehensive evaluation module is used for comprehensively evaluating a direct broadcast selling compliance index corresponding to the target sold commodity;
the direct broadcasting process processing module of the appointed ultrahigh-definition direct broadcasting room is used for carrying out targeted processing on the direct broadcasting process of the appointed ultrahigh-definition direct broadcasting room by the direct broadcasting platform according to the direct broadcasting selling compliance index corresponding to the target selling commodity;
the target selling commodity direct-broadcast selling conformity analysis module comprises a target selling commodity real object display conformity analysis unit, a target selling commodity description audio conformity analysis unit and a target selling commodity reply conformity analysis unit;
the target commodity selling object display conformity analysis unit is used for carrying out object display conformity analysis on an object display image corresponding to the target commodity selling object, and the specific analysis process is as follows:
clicking a shopping display link corresponding to a target commodity to extract detailed page appearance display information of the target commodity in the shopping display link, wherein the detailed page appearance display information comprises page display size specifications and page display color chroma;
performing page display outline delineation according to the page display size specification corresponding to the target sold commodity, and acquiring a page display outline surface area corresponding to the target sold commodity based on the page display outline delineation;
extracting a physical outline from a physical display image of the target sold commodity, acquiring a shooting proportion of the physical display image, and further carrying out scaling on the physical outline corresponding to the target sold commodity according to the shooting proportion in the same proportion to obtain a physical reference outline corresponding to the target sold commodity;
the method comprises the steps of carrying out coincidence comparison on a real object reference appearance outline corresponding to a target sold commodity and a page display appearance outline to obtain a coincidence area of the real object appearance outlines;
calculating the coincidence area of the outline of the object corresponding to the target commodity and the surface area of the display outline of the page through the coincidence coefficient of the display outline of the object
Obtaining a physical appearance display conforming coefficient corresponding to the target selling commodity;
extracting object color chroma from an object display image of the object commodity, comparing the object color chroma with page display color chroma corresponding to the object commodity, and displaying a coincidence coefficient calculation formula through the object color
Obtaining a color display fit coefficient of a real object corresponding to the target commodity;
calculating the real object display conformity degree corresponding to the target commodity according to the real object appearance display conformity coefficient and the real object color display conformity coefficient corresponding to the target commodity
The calculation formula is
Wherein A and B respectively represent proportion factors corresponding to the real object appearance display coincidence coefficient and the real object color display coincidence coefficient, and e represents a natural constant;
the target selling commodity description audio conformity analysis unit is used for carrying out description attribute conformity analysis on description audio information of the target selling commodity by the anchor, and the specific analysis process is as follows:
carrying out voice recognition on the description audio information of the target selling commodity by the anchor to obtain description text information of the target selling commodity by the anchor;
extracting description attribute information from description text information of the anchor on the target selling goods, and counting the number of the description attributes and specific information corresponding to various description attributes;
counting the number of display attributes and specific information corresponding to various display attributes from the attribute display information of the detail page of the target sold commodity in the shopping display link;
matching various description attributes corresponding to the target selling goods with various display attributes, counting the number of the description attributes successfully matched, and recording the description attributes successfully matched as matching description attributes;
comparing the specific information of each matching description attribute with the specific information of the corresponding display attribute, and recording the description coincidence coefficient corresponding to a certain matching description attribute as the specific information of the corresponding display attribute if the specific information of the matching description attribute is consistent with the specific information of the corresponding display attribute
If not, the description coincidence coefficient corresponding to the matching description attribute is recorded as
；
Extracting importance corresponding to various matching description attributes from a supervision service library based on the various matching description attributes corresponding to the target selling goods;
counting the corresponding description attribute conformity of the target selling commodity according to the number of the matching description attributes and the corresponding description conformity coefficient and the corresponding importance degree of each matching description attribute
The calculation formula is
K is expressed as the number of matching description attributes, K is expressed as the number of presentation attributes,
the description corresponding to the matching description attribute represented as the ith is consistent with the coefficient, i is represented as the number of the matching description attribute, i =1,2
Is taken as
Or
，
Expressing the importance corresponding to the ith matching description attribute;
the target selling commodity description audio conformity analysis unit is also used for carrying out description term conformity analysis on description audio information of the target selling commodity by the anchor, and the specific analysis process is as follows:
the method comprises the steps that the anchor carries out stop word removal and word segmentation processing on description text information of a target selling commodity to obtain a plurality of description word groups;
matching each description phrase with the constructed non-standard publicity phrase vocabulary library, counting the number of description phrases successfully matched according to the matching, and marking the number as the number of violation description phrases;
calculating the conformity of descriptive phrases corresponding to the target selling goods based on the number of descriptive phrases corresponding to the target selling goods and the number of illegal descriptive phrases
The calculation formula is
；
The reply conformity analysis unit of the target selling commodity is used for carrying out anchor reply conformity analysis on reply information of the challenge information by an anchor, and the specific analysis method is as follows:
counting the number of question attributes of the target selling goods from the question information;
extracting specific reply information of the anchor to various question attributes from public screen reply information of the anchor to the question information;
extracting specific display information corresponding to various question attributes from detail page attribute display information corresponding to the target selling commodity according to various question attributes of the watching user to the target selling commodity;
comparing the specific reply information of the anchor to various question attributes with the specific display information corresponding to the various question attributes, counting the number of the question attributes which are consistent in comparison, and recording the question attributes which are consistent in comparison as adaptive question attributes;
matching each adapted question attribute with the importance degree corresponding to each commodity attribute in the supervision service library, and matching the importance degree corresponding to each adapted question attribute from the importance degrees;
calculating the anchor reply conformity degree corresponding to the target selling commodity based on the question attribute quantity, the adaptive question attribute quantity and the importance degree corresponding to each adaptive question attribute of the watching user to the target selling commodity
In which
X represents the number of questioning attributes of the viewing user for the target offered merchandise,
the importance degree corresponding to the jth type of the adaptive question attributes is represented, j is represented as the number of the adaptive question attributes, and j =1, 2.
The direct broadcast selling compliance index evaluation formula corresponding to the target selling commodity is
，
Expressing as direct broadcast selling compliance index corresponding to target selling commodity, and respectively expressing as weight coefficients corresponding to set real object display conformity, description attribute conformity, description phrase conformity and anchor reply conformity;
the specific processing steps corresponding to the targeted processing of the live broadcast process of the designated ultra-high-definition live broadcast room are as follows:
comparing the direct broadcast selling compliance index corresponding to the target selling commodity with direct broadcast selling compliance index intervals corresponding to various direct broadcast selling compliance levels in the supervision service library, and screening out the direct broadcast selling compliance level corresponding to the target selling commodity;
and matching the live broadcast selling compliance grade corresponding to the target selling commodity with the live broadcast process processing modes corresponding to various live broadcast selling compliance grades in the supervision service library, so as to match the live broadcast process processing modes corresponding to the appointed ultra-high-definition live broadcast room, and executing the live broadcast process processing modes by a live broadcast platform.
2. The ultra high definition live system service supervision system based on data analysis according to claim 1, characterized in that: the display information comprises a real object display image and a shopping display link.
3. The ultra high definition live system service supervision system based on data analysis according to claim 1, characterized in that: the live broadcast process processing modes corresponding to various live broadcast selling compliance levels comprise normal live broadcast continuing, live broadcast interruption prompting and live broadcast permission blocking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211002379.0A CN115086721B (en) | 2022-08-22 | 2022-08-22 | Ultra-high-definition live system service supervision system based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211002379.0A CN115086721B (en) | 2022-08-22 | 2022-08-22 | Ultra-high-definition live system service supervision system based on data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115086721A CN115086721A (en) | 2022-09-20 |
| CN115086721B true CN115086721B (en) | 2022-10-25 |
Family
ID=83244106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211002379.0A Active CN115086721B (en) | 2022-08-22 | 2022-08-22 | Ultra-high-definition live system service supervision system based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115086721B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738290A (en) * | 2020-05-14 | 2020-10-02 | 北京沃东天骏信息技术有限公司 | Image detection method, model construction and training method, device, equipment and medium |
| CN111770353A (en) * | 2020-06-24 | 2020-10-13 | 北京字节跳动网络技术有限公司 | Live broadcast monitoring method and device, electronic equipment and storage medium |
| CN112492343A (en) * | 2020-12-16 | 2021-03-12 | 浙江大华技术股份有限公司 | Video live broadcast monitoring method and related device |
| CN112995696A (en) * | 2021-04-20 | 2021-06-18 | 共道网络科技有限公司 | Live broadcast room violation detection method and device |
| CN113315994A (en) * | 2021-04-23 | 2021-08-27 | 北京达佳互联信息技术有限公司 | Live broadcast data processing method and device, electronic equipment and storage medium |
| US11172258B1 (en) * | 2021-01-26 | 2021-11-09 | NortonLifeLock Inc. | Protecting against an impersonation scam in a live video stream |
| CN114579747A (en) * | 2022-03-10 | 2022-06-03 | 中国计量大学 | Method for recognizing illegal propaganda of direct-broadcast marketing of cosmetics |
| CN114786035A (en) * | 2022-05-25 | 2022-07-22 | 上海氪信信息技术有限公司 | Compliance quality inspection and interactive question-answering system and method for live scene |
-
2022
- 2022-08-22 CN CN202211002379.0A patent/CN115086721B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738290A (en) * | 2020-05-14 | 2020-10-02 | 北京沃东天骏信息技术有限公司 | Image detection method, model construction and training method, device, equipment and medium |
| CN111770353A (en) * | 2020-06-24 | 2020-10-13 | 北京字节跳动网络技术有限公司 | Live broadcast monitoring method and device, electronic equipment and storage medium |
| CN112492343A (en) * | 2020-12-16 | 2021-03-12 | 浙江大华技术股份有限公司 | Video live broadcast monitoring method and related device |
| US11172258B1 (en) * | 2021-01-26 | 2021-11-09 | NortonLifeLock Inc. | Protecting against an impersonation scam in a live video stream |
| CN112995696A (en) * | 2021-04-20 | 2021-06-18 | 共道网络科技有限公司 | Live broadcast room violation detection method and device |
| CN113315994A (en) * | 2021-04-23 | 2021-08-27 | 北京达佳互联信息技术有限公司 | Live broadcast data processing method and device, electronic equipment and storage medium |
| CN114579747A (en) * | 2022-03-10 | 2022-06-03 | 中国计量大学 | Method for recognizing illegal propaganda of direct-broadcast marketing of cosmetics |
| CN114786035A (en) * | 2022-05-25 | 2022-07-22 | 上海氪信信息技术有限公司 | Compliance quality inspection and interactive question-answering system and method for live scene |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115086721A (en) | 2022-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cheng et al. | Multimedia features for click prediction of new ads in display advertising | |
| CN103729785B (en) | Video user gender classification method and device for method | |
| US20070250856A1 (en) | Distinguishing National and Local Broadcast Advertising and Other Content | |
| Zhang et al. | Fighting strategies in a market with counterfeits | |
| US20100095326A1 (en) | Program content tagging system | |
| CN106611344A (en) | Method and device for mining potential customers | |
| US20100232644A1 (en) | System and method for counting the number of people | |
| CN101459806A (en) | System and method for video playing | |
| CN111832296A (en) | Multimedia advertisement intelligent delivery management system based on big data | |
| CN109615440B (en) | Direct reaction advertisement detection and classification system and method | |
| CN111667337A (en) | Commodity evaluation ordering method and system | |
| US10779036B1 (en) | Automated identification of product or brand-related metadata candidates for a commercial using consistency between audio and image elements of products or brands detected in commercials | |
| Cheng et al. | Everyone can be a star: Understanding the role of live video streaming in online retail | |
| CN115086721B (en) | Ultra-high-definition live system service supervision system based on data analysis | |
| CN109034101B (en) | One-to-many dynamic and static advertisement playing method | |
| EP3776431A1 (en) | Automated identification of product or brand-related metadata candidates for a commercial using persistence of product or brand-related text or objects in video frames of the commercial | |
| CN113256356A (en) | Internet advertisement intelligent delivery analysis management system based on feature recognition | |
| CN111709762A (en) | Information matching degree evaluation method, device, equipment and storage medium | |
| KR102404247B1 (en) | Customer management system | |
| US20210342856A1 (en) | Customer service assistance apparatus, customer service assistance method, and computer-readable recording medium | |
| US11410216B2 (en) | Customer service assistance apparatus, customer service assistance method, and computer-readable recording medium | |
| CN114328490A (en) | Construction method, equipment and medium for live broadcast e-commerce analysis display | |
| CN115392993A (en) | Big data intelligent management system based on live video commodity storage show | |
| CN113962765A (en) | Retail intelligent interactive exhibition and sale system | |
| CN112580674A (en) | Picture identification method, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |