CN113507623B - Intelligent commodity selection system and method based on data analysis - Google Patents
Intelligent commodity selection system and method based on data analysis Download PDFInfo
- Publication number
- CN113507623B CN113507623B CN202110774854.5A CN202110774854A CN113507623B CN 113507623 B CN113507623 B CN 113507623B CN 202110774854 A CN202110774854 A CN 202110774854A CN 113507623 B CN113507623 B CN 113507623B
- Authority
- CN
- China
- Prior art keywords
- commodity
- live broadcast
- candidate
- module
- value
- 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
- H04N21/21—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
- H04N21/25—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
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44213—Monitoring of end-user related data
- H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
-
- 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
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
-
- 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
- H04N21/47815—Electronic shopping
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Computer Networks & Wireless Communication (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent commodity selection system and method based on data analysis, wherein the intelligent commodity selection system comprises a to-be-selected information acquisition module, a candidate commodity selection module, a preferred commodity selection module and a preferred commodity pushing module, the to-be-selected information acquisition module is used for acquiring information of each to-be-selected commodity which is subjected to direct broadcast, the number of the to-be-selected commodities is a, the candidate commodity selection module is used for acquiring the selection condition of each direct broadcast in a latest preset time period and selecting b candidate commodities from the to-be-selected commodities, the preferred commodity selection module is used for acquiring the sales condition of the selected commodities in each direct broadcast in the latest preset time period and selecting c preferred commodities from the candidate commodities, the preferred commodity pushing module is used for pushing the preferred commodities to a worker, wherein a, b and c are natural numbers, and a is larger than b and larger than c.
Description
Technical Field
The invention relates to the technical field of commodity selection, in particular to a commodity intelligent selection system and method based on data analysis.
Background
With the development of internet technology, live broadcast is more and more popular. Live e-commerce sales are becoming popular due to the wider audience available for live broadcast. Live electronic commerce selling is different from traditional electronic commerce selling, live electronic commerce selling is a form of an anchor chat with numerous audiences through screens, entertainment and interactivity of the live electronic commerce selling are higher than those of the traditional electronic commerce selling, shopping guide stimulation degree of consumers can be enhanced, and memory degree of people to commodities is deepened. The anchor can select from a large pile of commodities before starting to live and sell the commodities, in the prior art, the commodities are selected mainly in a manual selection mode, but the mode needs to consume large manpower.
Disclosure of Invention
The invention aims to provide a commodity intelligent selection system and a commodity intelligent selection method based on data analysis, so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme: the intelligent commodity selection system based on data analysis comprises a to-be-selected information acquisition module, a candidate commodity selection module, a preferred commodity selection module and a preferred commodity pushing module, wherein the to-be-selected information acquisition module is used for acquiring information of each to-be-selected commodity which is subjected to direct broadcast, the number of the to-be-selected commodities is a, the candidate commodity selection module is used for acquiring selection conditions of each direct broadcast within a latest preset time period and selecting b candidate commodities from the to-be-selected commodities, the preferred commodity selection module is used for acquiring selection sales conditions of each direct broadcast within the latest preset time period and selecting c preferred commodities from the candidate commodities, and the preferred commodity pushing module is used for pushing the preferred commodities to a worker, wherein a, b and c are natural numbers, and a is larger than b and larger than c.
Further, the candidate commodity selection module includes a related commodity value acquisition module, a time commodity value acquisition module and a pre-selected value calculation sorting module, where the number of times of live broadcast in the latest period of time acquired by the related commodity value acquisition module is Nz, the number of times that a selected commodity is the same as the type of a to-be-selected commodity exists in each live broadcast is Nx, and a selected commodity that is the same as the type of the to-be-selected commodity is set as the related commodity of the to-be-selected commodity, then a related commodity value U of a certain to-be-selected commodity is Nx/Nz, the time commodity value acquisition module acquires a time interval T between the live broadcast time of the related commodity in which the to-be-selected commodity appears last time and the live broadcast time, and performs normalization processing on the time interval corresponding to each to-be-selected commodity, then a time commodity value V of a certain to-be-selected commodity is (T-Tmin)/(Tmax-Tmin), where, tmax is the maximum value of the time interval corresponding to each commodity to be selected, Tmin is the minimum value of the time interval corresponding to each commodity to be selected, the preselected value calculating and sorting module calculates the preselected value of each commodity to be selected according to the associated selected value and the time selected value, sorts the preselected values in descending order, and selects the commodity to be selected before sorting as a candidate commodity.
Further, the preferred commodity selection module includes a first parameter obtaining module, a second parameter obtaining module, a third parameter obtaining module, and a comprehensive selection value calculating and sorting module, where the first parameter obtaining module counts an average value X of a ratio between an actual sales volume and a supply volume of an associated commodity of each candidate commodity, and performs normalization processing on the average value X corresponding to each candidate commodity, so that a first parameter H of a certain candidate commodity is (X-Xmin)/(Xmax-Xmin), where Xmax is a maximum value of the average value X corresponding to each candidate commodity, Xmin is a minimum value of the average value X corresponding to each candidate commodity, the second parameter obtaining module obtains an average value Y of a ratio between a number of comments related to the associated commodity and a total number of comments received when the associated commodity is live broadcast when the associated commodity of each candidate commodity is live broadcast, normalizing the average value X corresponding to each candidate product, so that the second parameter I of a certain candidate product is (Y-Ymin)/(Ymax-Ymin), wherein Ymax is the maximum value of the average value Y corresponding to each candidate commodity, Ymin is the minimum value of the average value Y corresponding to each candidate commodity, the third parameter acquisition module acquires the number of watching persons in each live broadcast in the latest preset time period, acquires the stability index of the number of the watching persons when the associated commodities of each candidate commodity are live broadcast, calculates the average value corresponding to each stability index as a third parameter Z, the comprehensive selection value calculating and sorting module calculates comprehensive selection values of the candidate commodities according to the first parameter, the second parameter and the third parameter, sorts the comprehensive selection values of the candidate commodities in a descending order, and takes the candidate commodity c before sorting as a preferred commodity.
Further, the third parameter obtaining module comprises a stability index obtaining module and a third parameter calculating module, the stability index obtaining module obtains the stability index of the number of people watching the related commodities of each candidate commodity during live broadcasting in the latest preset time period, the third parameter calculating module calculates the average value corresponding to each stability index to be a third parameter Z, the stability index obtaining module comprises a static user collecting and marking module, a dynamic user marking module, a user correcting, collecting and judging module and a stability index calculating module, the static user collecting and marking module marks the live watching user before starting live broadcasting a certain related commodity as a static user and collects the number Rz of the static user, the dynamic user marking module collects the number Rz of the static user in the preset time after the related commodity is live broadcast, if detecting that a certain initial static user quits watching the live broadcast, the static user is marked as a dynamic user, the user correction acquisition judging module acquires a time node when the dynamic user reenters to watch live broadcast as an investigation node, if the investigation node is located in the live broadcast process of the associated commodity, the time node when the dynamic user quits to watch live broadcast next time is acquired as an auxiliary node T1, and under the condition that the auxiliary node T1 is located in the live broadcast process of the associated commodity, a reference Tk is T1-T2, wherein T2 is the time node when the live broadcast of the associated commodity is finished, and when the reference Tk is greater than a reference threshold, the dynamic user is modified as a static user, and the stability index calculation module calculates the stability index Rx (Rd/Rz) of the associated commodity according to the number of the static users Rd when the live broadcast of the associated commodity is finished.
An intelligent commodity selection method based on data analysis comprises the following steps:
obtaining the information of each commodity to be selected of the live broadcast, wherein the number of the commodities to be selected is a,
collecting the selection condition of each live broadcast in the latest preset time period, selecting b candidate commodities from the commodities to be selected,
collecting the selected goods sale condition in each live broadcast in the latest preset time period, selecting c preferable goods from the candidate goods, and pushing the preferable goods to workers, wherein a, b and c are natural numbers, and a is larger than b and larger than c.
Further, the selecting the candidate goods from the goods to be selected includes:
acquiring the number of live broadcast times Nz in the latest period of time, and the number of times that the type of the selected product is the same as that of the to-be-selected commodity exists in each live broadcast as Nx, and setting the selected product with the same type as that of the to-be-selected commodity as a related commodity of the to-be-selected commodity, so that the related selected value U of a certain to-be-selected commodity is Nx/Nz;
acquiring the time interval T between the live broadcast time of the associated commodity of the latest commodity to be selected and the live broadcast time, and performing normalization processing on the time interval corresponding to each commodity to be selected, wherein the time selection value V of a certain commodity to be selected is (T-Tmin)/(Tmax-Tmin), Tmax is the maximum value of the time interval corresponding to each commodity to be selected, Tmin is the minimum value of the time interval corresponding to each commodity to be selected,
then the pre-selected value P of a certain candidate item is 0.6U + 0.4V,
and (4) sorting the preselected values in a descending order, and selecting the commodities to be selected before the sorting as candidate commodities.
Further, the selecting a preferred commodity from the candidate commodities comprises:
the average value X of the ratio of the actual sales volume to the supply volume of the associated commodity of each candidate commodity is counted, and the average value X corresponding to each candidate commodity is normalized, so that the first parameter of a certain candidate commodity
H ═ X-Xmin)/(Xmax-Xmin), where Xmax is the maximum value of the average value X corresponding to each candidate commodity, and Xmin is the minimum value of the average value X corresponding to each candidate commodity;
acquiring an average value Y of the ratio of the number of comments related to the associated commodity to the total number of comments received when the associated commodity is live broadcast when the associated commodity of each candidate commodity is live broadcast, and normalizing the average value X corresponding to each candidate commodity, wherein a second parameter I of a certain candidate commodity is (Y-Ymin)/(Ymax-Ymin), Ymax is the maximum value of the average value Y corresponding to each candidate commodity, and Ymin is the minimum value of the average value Y corresponding to each candidate commodity;
acquiring a stability index of the number of viewers in each live broadcast of the associated commodities of each candidate commodity in the latest preset time period, and calculating a corresponding average value of each stability index as a third parameter Z;
then the comprehensive selected value P of a certain candidate commodity is 0.38X + 0.26Y + 0.3Z;
and (4) sorting the comprehensive selection values in a descending order, and taking the candidate commodity c before sorting as the preferred commodity.
Further, obtaining the number of viewers who are watching the associated commodities of each candidate commodity on-demand comprises:
marking the live broadcast watching users before starting live broadcast of a certain associated commodity as static users, collecting the number Rz of the static users,
within a preset time after the associated commodity is live broadcast, if an initial static user is detected to quit watching the live broadcast, the static user is marked as a dynamic user,
the time node for collecting the dynamic user to re-enter and watch the live broadcast is an investigation node, if the investigation node is positioned in the live broadcast process of the associated commodity, the time node for collecting the dynamic user to quit and watch the live broadcast next time is an auxiliary node T1,
if the auxiliary node T1 is located in the live broadcast process of the associated merchandise, the reference Tk is T1-T2, where T2 is the time node when the live broadcast of the associated merchandise is finished
When the reference Tk is larger than the reference threshold, modifying the dynamic user into a static user;
and counting the number Rd of static users at the end of the live broadcast of the associated commodity, wherein the stability index Rx of the associated commodity is Rd/Rz.
Compared with the prior art, the invention has the following beneficial effects: according to the method and the device, the candidate commodities are selected from the commodities to be selected according to the selection condition of each live broadcast in the latest preset time period, and then the preferred commodities are selected from the candidate commodities, so that the comparison calculation amount in the commodity selection process is reduced, and the selected preferred commodities are recommended to the staff, so that the selection comparison burden of the staff is reduced.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic block diagram of a commodity intelligent selection system based on data analysis according to 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a technical solution: the intelligent commodity selection system based on data analysis comprises a to-be-selected information acquisition module, a candidate commodity selection module, a preferred commodity selection module and a preferred commodity pushing module, wherein the to-be-selected information acquisition module is used for acquiring information of each to-be-selected commodity which is subjected to direct broadcast, the number of the to-be-selected commodities is a, the candidate commodity selection module is used for acquiring selection conditions of each direct broadcast within a latest preset time period and selecting b candidate commodities from the to-be-selected commodities, the preferred commodity selection module is used for acquiring selection sales conditions of each direct broadcast within the latest preset time period and selecting c preferred commodities from the candidate commodities, and the preferred commodity pushing module is used for pushing the preferred commodities to a worker, wherein a, b and c are natural numbers, and a is larger than b and larger than c.
The candidate commodity selection module comprises a correlation commodity value acquisition module, a time commodity value acquisition module and a preselected value calculation sorting module, wherein the correlation commodity value acquisition module acquires that the number of times of live broadcast in a latest period of time is Nz, the number of times of live broadcast of a commodity which is the same as the type of the to-be-selected commodity exists in each live broadcast is Nx, a commodity which is the same as the type of the to-be-selected commodity is set as the correlated commodity of the to-be-selected commodity, then a correlation commodity value U of a certain to-be-selected commodity is Nx/Nz, the time commodity value acquisition module acquires a time interval T between the live broadcast time of the correlated commodity which the to-be-selected commodity appears last time and the live broadcast time, and carries out normalization processing on the time interval corresponding to each to-be-selected commodity, then a time commodity value V of the certain to-be-selected commodity is (T-Tmin)/(Tmax-Tmin), wherein Tmax is the maximum value of the time interval corresponding to each to-selected commodity, tmin is the minimum value of the time interval corresponding to each commodity to be selected, the preselected value calculating and sorting module calculates the preselected value of each commodity to be selected according to the associated selected value and the time selected value, sorts the preselected values in descending order, and selects the commodity to be selected before sorting as a candidate commodity.
The preferred commodity selection module comprises a first parameter acquisition module, a second parameter acquisition module, a third parameter acquisition module and a comprehensive selection value calculation sorting module, wherein the first parameter acquisition module counts an average value X of the ratio of actual sales volume to supply volume of the associated commodities of each candidate commodity, and normalizes the average value X corresponding to each candidate commodity, so that a first parameter H of a certain candidate commodity is (X-Xmin)/(Xmax-Xmin), wherein Xmax is the maximum value of the average value X corresponding to each candidate commodity, Xmin is the minimum value of the average value X corresponding to each candidate commodity, the second parameter acquisition module acquires the average value Y of the ratio of the number of comments related to the associated commodities to the same commodities received when the same commodities are directly broadcast, normalizing the average value X corresponding to each candidate product, so that the second parameter I of a certain candidate product is (Y-Ymin)/(Ymax-Ymin), wherein Ymax is the maximum value of the average value Y corresponding to each candidate commodity, Ymin is the minimum value of the average value Y corresponding to each candidate commodity, the third parameter acquisition module acquires the number of watching persons in each live broadcast in the latest preset time period, acquires the stability index of the number of the watching persons when the associated commodities of each candidate commodity are live broadcast, calculates the average value corresponding to each stability index as a third parameter Z, the comprehensive selection value calculating and sorting module calculates comprehensive selection values of the candidate commodities according to the first parameter, the second parameter and the third parameter, sorts the comprehensive selection values of the candidate commodities in a descending order, and takes the candidate commodity c before sorting as a preferred commodity.
The third parameter obtaining module comprises a stability index obtaining module and a third parameter calculating module, the stability index obtaining module is used for obtaining the stability index of the number of watching persons during the live broadcast of the associated commodities of each candidate commodity in the latest preset time period, the third parameter calculating module is used for calculating the average value corresponding to each stability index to be a third parameter Z, the stability index obtaining module comprises a static user collecting and marking module, a dynamic user marking module, a user correcting, collecting and judging module and a stability index calculating module, the static user collecting and marking module is used for marking the live watching user before the live broadcast of a certain associated commodity as a static user and collecting the number Rz of the static users, the dynamic user marking module is used for marking the live watching user before the live broadcast of the certain associated commodity as a static user and collecting the number Rz of the static users within the preset time after the live broadcast of the associated commodity if detecting that a certain initial static user quits watching the live broadcast, the static user is marked as a dynamic user, the user correction acquisition judging module acquires a time node when the dynamic user reenters to watch live broadcast as an investigation node, if the investigation node is located in the live broadcast process of the associated commodity, the time node when the dynamic user quits to watch live broadcast next time is acquired as an auxiliary node T1, and under the condition that the auxiliary node T1 is located in the live broadcast process of the associated commodity, a reference Tk is T1-T2, wherein T2 is the time node when the live broadcast of the associated commodity is finished, and when the reference Tk is greater than a reference threshold, the dynamic user is modified as a static user, and the stability index calculation module calculates the stability index Rx (Rd/Rz) of the associated commodity according to the number of the static users Rd when the live broadcast of the associated commodity is finished.
An intelligent commodity selection method based on data analysis comprises the following steps:
acquiring information of each to-be-selected commodity of the live broadcast, wherein the number of the to-be-selected commodities is a, and the information of the to-be-selected commodities is mainly the type of the commodity;
collecting the selection condition of each live broadcast in the latest preset time period, and selecting b candidate commodities from the commodities to be selected, wherein the selection of the candidate commodities from the commodities to be selected comprises the following steps:
acquiring the number of live broadcast times Nz in the latest period of time, and the number of times that the type of the selected product is the same as that of the to-be-selected commodity exists in each live broadcast as Nx, and setting the selected product with the same type as that of the to-be-selected commodity as a related commodity of the to-be-selected commodity, so that the related selected value U of a certain to-be-selected commodity is Nx/Nz; for example, a certain selected commodity is a nut of xxx brand, the type of the selected commodity is a nut, five live broadcast times are provided in the latest period of time, no nut is provided in the first live broadcast, xxxx brand nuts are provided in the second live broadcast and the third live broadcast, no nut is provided in the fourth live broadcast and the fifth live broadcast, Nz is 5, Nx is 2, U is 2/5, xxxx brand nuts are related commodities of which the selected commodity is a nut of xxx brand
Acquiring the time interval T between the live broadcast time of the associated commodity of the latest commodity to be selected and the live broadcast time, and performing normalization processing on the time interval corresponding to each commodity to be selected, wherein the time selection value V of a certain commodity to be selected is (T-Tmin)/(Tmax-Tmin), Tmax is the maximum value of the time interval corresponding to each commodity to be selected, Tmin is the minimum value of the time interval corresponding to each commodity to be selected,
in the application, a time selection value is obtained based on the condition that each live broadcast time interval is equal, if the live broadcast time intervals are different, the time selection value V of a certain to-be-selected commodity is (G-Gmin)/(Gmax-Gmin) for obtaining the live broadcast time interval G between the live broadcast time and the live broadcast time of the relevant commodity of the to-be-selected commodity which occurs the latest time, and performing normalization processing on the time interval corresponding to each to-be-selected commodity, wherein Gmax is the maximum value of the live broadcast time interval corresponding to each to-be-selected commodity, Gmin is the minimum value of the live broadcast time interval corresponding to each to-be-selected commodity, for example, in the above case, five live broadcasts are performed in total within the latest period of time, no nut is subjected to the first live broadcast, no nut is subjected to the second live broadcast, no nut with xxxx brand is subjected to the third live broadcast, no nut is subjected to the fourth live broadcast, no nut is subjected to the fifth live broadcast, then G-2 for xxx-brand nuts
Then the pre-selected value P of a certain candidate item is 0.6U + 0.4V,
sorting the preselected values in a descending order, and selecting the commodities to be selected before sorting as candidate commodities; when the associated option value of a certain to-be-selected commodity is larger, the user is more favored to select the commodity of the type, and when the time option value of the certain to-be-selected commodity is smaller, the user is easy to feel tired when the commodity of the type just comes up;
collecting the selected goods sale condition in each live broadcast in the latest preset time period, selecting c preferable goods from the candidate goods, and pushing the preferable goods to workers, wherein a, b and c are natural numbers, and a is larger than b and larger than c.
The selecting of the preferred goods from the candidate goods comprises:
the average value X of the ratio of the actual sales volume to the supply volume of the associated commodity of each candidate commodity is counted, and the average value X corresponding to each candidate commodity is normalized, so that the first parameter of a certain candidate commodity
H ═ X-Xmin)/(Xmax-Xmin), where Xmax is the maximum value of the average value X corresponding to each candidate commodity, and Xmin is the minimum value of the average value X corresponding to each candidate commodity; the greater the ratio of sales to supply, the more popular this type of merchandise is;
acquiring an average value Y of the ratio of the number of comments related to the associated commodity to the total number of comments received when the associated commodity is live broadcast when the associated commodity of each candidate commodity is live broadcast, and normalizing the average value X corresponding to each candidate commodity, wherein a second parameter I of a certain candidate commodity is (Y-Ymin)/(Ymax-Ymin), Ymax is the maximum value of the average value Y corresponding to each candidate commodity, and Ymin is the minimum value of the average value Y corresponding to each candidate commodity; for example, if the to-be-selected commodity is a xxx-brand nut, in the past related commodities, as long as the number of comments related to the nuts in the time period of the second time live broadcast and the third time live broadcast of the nuts with brands appears, if the number of all comments in the time period of the second time live broadcast of the xxxx-brand nuts is f1, the number of comments related to the nuts is e1, the number of all comments in the time period of the third time live broadcast of the xxxx-brand nuts is f2, and the number of comments related to the nuts is e2, then the average value Y corresponding to the age of the xxx-brand nuts is (e1/f1+ e2/f 1)/2; the larger the average value Y is, the higher the attention degree of the candidate commodity is;
acquiring the stable index of the number of watching persons in the direct broadcasting of the associated commodities of each candidate commodity in each direct broadcasting in the latest preset time period,
the obtaining of the number of viewers in the live broadcast of the associated commodities of the candidate commodities comprises:
marking a live broadcast watching user before starting live broadcast of a certain associated commodity as a static user, and acquiring the number Rz of the static user, wherein if the last live broadcast commodity is milk and the commodity to be live broadcast is nuts, the acquisition of the live broadcast milk is finished, and the live broadcast watching user before starting live broadcast of the nuts is the static user;
within a preset time after the associated commodity is live broadcast, if a certain static user is detected to quit watching the live broadcast, marking the static user as a dynamic user, such as within a preset time period of live broadcasting nuts, and if a certain static user is detected to quit watching the live broadcast of nut sales, marking the static user as a dynamic user,
the time node for collecting the dynamic user to re-enter and watch the live broadcast is an investigation node, if the investigation node is positioned in the live broadcast process of the associated commodity, the time node for collecting the dynamic user to quit and watch the live broadcast next time is an auxiliary node T1,
if the auxiliary node T1 is located in the live broadcast process of the associated merchandise, the reference Tk is T1-T2, where T2 is the time node when the live broadcast of the associated merchandise is finished
When the reference Tk is larger than the reference threshold, modifying the dynamic user into a static user;
counting the number Rd of static users when the live broadcast of the associated commodity is finished, wherein the stability index Rx of the associated commodity is Rd/Rz; the static user refers to a static user after judging whether to modify the dynamic user into the static user;
calculating the average value corresponding to each stability index as a third parameter Z; the situation that a user quits in the middle of watching the live broadcast often occurs in the live broadcast, some users quit because the user has no interest in the live broadcast commodity, some users quit because other things exist, whether the user quits to watch the live broadcast because other things exist is judged according to the in-out situation and the in-out time node of the commodity watching process of the user, and therefore the attraction situation of the commodity to the user can be judged more accurately.
Then the comprehensive selected value P of a certain candidate commodity is 0.38X + 0.26Y + 0.3Z;
and (4) sorting the comprehensive selection values in a descending order, and taking the candidate commodity c before sorting as the preferred commodity.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (2)
1. An intelligent commodity selection system based on data analysis is characterized by comprising a to-be-selected information acquisition module, a candidate commodity selection module, a preferred commodity selection module and a preferred commodity pushing module, wherein the to-be-selected information acquisition module is used for acquiring information of each to-be-selected commodity which is live broadcast, the number of the to-be-selected commodities is a, the candidate commodity selection module is used for acquiring the selection condition of each live broadcast in a latest preset time period and selecting b candidate commodities from the to-be-selected commodities, the preferred commodity selection module is used for acquiring the sales condition of the selected commodities in each live broadcast in the latest preset time period and selecting c preferred commodities from the candidate commodities, the preferred commodity pushing module is used for pushing the preferred commodities to a worker, a, b and c are natural numbers, and a is larger than b and larger than c;
the candidate commodity selecting module comprises a related commodity value obtaining module, a time commodity value obtaining module and a preselected value calculating and sorting module, wherein the related commodity value obtaining module is used for collecting the number of times of live broadcast in the latest period of time and collecting the number of times of live broadcast of a commodity to be selected as Nz, the number of times of live broadcast of the commodity to be selected as Nx exists in each live broadcast, the commodity of the same type as the commodity to be selected is set as the related commodity of the commodity to be selected, the related commodity value U = Nx/Nz of a certain commodity to be selected, the time commodity value obtaining module is used for obtaining the time interval T between the live broadcast time of the related commodity of the commodity to be selected which occurs the latest time and the live broadcast time, the time interval corresponding to each commodity to be selected is normalized, the time commodity value V = (T-Tmin)/(Tmax-Tmin) of the certain commodity to be selected is used for calculating and sorting module, tmin is the minimum value of the time interval corresponding to each commodity to be selected, the preselected value calculating and sorting module calculates the preselected value P = 0.6U + 0.4V of each commodity to be selected according to the associated selected value and the time selected value, sorts the preselected values in descending order, and selects the commodity to be selected before sorting b as a candidate commodity;
the preferred commodity selection module comprises a first parameter acquisition module, a second parameter acquisition module, a third parameter acquisition module and a comprehensive selection value calculation sorting module, wherein the first parameter acquisition module counts an average value X of the ratio of actual sales volume to supply volume of the associated commodity of each candidate commodity and normalizes the average value X corresponding to each candidate commodity, so that a first parameter H = (X-Xmin)/(Xmax-Xmin) of a certain candidate commodity is obtained, wherein Xmax is the maximum value of the average value X corresponding to each candidate commodity, Xmin is the minimum value of the average value X corresponding to each candidate commodity, the second parameter acquisition module acquires the average value Y of the ratio of the number of comments related to the associated commodity to the number of the total comments received when the associated commodity is directly broadcast, the average value Y corresponding to each candidate commodity is normalized, and then the second parameter I = (Y-Ymin)/(Ymax-Ymin) for a certain candidate commodity, wherein Ymax is the maximum value of the average value Y corresponding to each candidate commodity, Ymin is the minimum value of the average value Y corresponding to each candidate commodity, the third parameter acquisition module acquires the number of watching persons in each live broadcast in the latest preset time period, acquires the stability index of the number of the watching persons when the associated commodities of each candidate commodity are live broadcast, calculates the average value corresponding to each stability index as a third parameter Z, the comprehensive selected value calculation and sorting module calculates the comprehensive selected value P = 0.38X + 0.26Y + 0.3Z of each candidate commodity according to the first parameter, the second parameter and the third parameter, sorts the comprehensive selected values of each candidate commodity according to the sequence from large to small, and takes the candidate commodity before sorting as the preferred commodity;
the third parameter obtaining module comprises a stability index obtaining module and a third parameter calculating module, the stability index obtaining module is used for obtaining the stability index of the number of watching persons during the live broadcast of the associated commodities of each candidate commodity in the latest preset time period, the third parameter calculating module is used for calculating the average value corresponding to each stability index to be a third parameter Z, the stability index obtaining module comprises a static user collecting and marking module, a dynamic user marking module, a user correcting, collecting and judging module and a stability index calculating module, the static user collecting and marking module is used for marking the live watching user before the live broadcast of a certain associated commodity as a static user and collecting the number Rz of the static users, the dynamic user marking module is used for marking the live watching user before the live broadcast of the certain associated commodity as a static user and collecting the number Rz of the static users within the preset time after the live broadcast of the associated commodity if detecting that a certain initial static user quits watching the live broadcast, the static user is marked as a dynamic user, the user correction acquisition judging module acquires that a time node when the dynamic user reenters to watch live broadcast is an investigation node, if the investigation node is located in the live broadcast process of the associated commodity, the time node when the dynamic user quits to watch live broadcast next time is acquired as an auxiliary node T1, and under the condition that the auxiliary node T1 is located in the live broadcast process of the associated commodity, a reference quantity Tk = T1-T2 is set, wherein T2 is the time node when the live broadcast of the associated commodity is finished, when the reference quantity Tk is larger than a reference quantity threshold value, the dynamic user is modified to be a static user, and the stability index calculating module calculates the stability index Rx = Rd/Rz of the associated commodity according to the number Rd of the static user when the live broadcast of the associated commodity is finished.
2. An intelligent commodity selection method based on data analysis is characterized in that: the intelligent product selection method comprises the following steps:
obtaining the information of each commodity to be selected of the live broadcast, wherein the number of the commodities to be selected is a,
collecting the selection condition of each live broadcast in the latest preset time period, selecting b candidate commodities from the commodities to be selected,
collecting the sales condition of selected goods in each live broadcast in a latest preset time period, selecting c preferred goods from the candidate goods, and pushing the preferred goods to workers, wherein a, b and c are natural numbers, and a is larger than b and larger than c;
the selecting candidate commodities from the commodities to be selected comprises the following steps:
acquiring the number of live broadcast times Nz in the last period of time, acquiring the number of times Nx of the selected item with the same type as the to-be-selected commodity in each live broadcast, and setting the selected item with the same type as the to-be-selected commodity as a related commodity of the to-be-selected commodity, wherein the related selected value U = Nx/Nz of a certain to-be-selected commodity;
acquiring the time interval T between the live broadcast time of the associated commodity of the latest commodity to be selected and the live broadcast time, and performing normalization processing on the time interval corresponding to each commodity to be selected, wherein a time option value V = (T-Tmin)/(Tmax-Tmin) of a certain commodity to be selected is obtained, Tmax is the maximum value of the time interval corresponding to each commodity to be selected, Tmin is the minimum value of the time interval corresponding to each commodity to be selected,
then the preselected value P = 0.6U + 0.4V for a certain commodity to be selected,
sorting the preselected values in a descending order, and selecting the commodities to be selected before sorting as candidate commodities;
the selecting of the preferred goods from the candidate goods comprises:
counting an average value X of the ratio of actual sales volume to supply volume of the associated commodity of each candidate commodity, and performing normalization processing on the average value X corresponding to each candidate commodity, wherein a first parameter H = (X-Xmin)/(Xmax-Xmin) of a certain candidate commodity is obtained, wherein Xmax is the maximum value of the average value X corresponding to each candidate commodity, and Xmin is the minimum value of the average value X corresponding to each candidate commodity;
acquiring an average value Y of the ratio of the number of comments related to the associated commodity to the total number of comments received when the associated commodity is live broadcast when the associated commodity of each candidate commodity is live broadcast, and normalizing the average value Y corresponding to each candidate commodity, wherein a second parameter I = (Y-Ymin)/(Ymax-Ymin) of a certain candidate commodity is obtained, Ymax is the maximum value of the average value Y corresponding to each candidate commodity, and Ymin is the minimum value of the average value Y corresponding to each candidate commodity;
acquiring a stability index of the number of viewers in each live broadcast of the associated commodities of each candidate commodity in the latest preset time period, and calculating a corresponding average value of each stability index as a third parameter Z;
then the integrated pick value P = 0.38X + 0.26Y + 0.3Z for a candidate good;
sorting the comprehensive selection values in a descending order, and taking the candidate commodity c before sorting as a preferred commodity;
the obtaining of the number of viewers in the live broadcast of the associated commodities of the candidate commodities comprises:
marking the live broadcast watching users before starting live broadcast of a certain associated commodity as static users, collecting the number Rz of the static users,
within a preset time after the associated commodity is live broadcast, if an initial static user is detected to quit watching the live broadcast, the static user is marked as a dynamic user,
the time node for collecting the dynamic user to re-enter and watch the live broadcast is an investigation node, if the investigation node is positioned in the live broadcast process of the associated commodity, the time node for collecting the dynamic user to quit and watch the live broadcast next time is an auxiliary node T1,
if the auxiliary node T1 is located in the live broadcast process of the associated merchandise, the reference quantity Tk = T1-T2, wherein T2 is the time node when the live broadcast of the associated merchandise is finished
When the reference quantity Tk is larger than the reference quantity threshold value, modifying the dynamic user into a static user;
and counting the number Rd of static users at the end of the live broadcast of the associated commodity, wherein the stability index Rx = Rd/Rz of the associated commodity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110774854.5A CN113507623B (en) | 2021-07-09 | 2021-07-09 | Intelligent commodity selection system and method based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110774854.5A CN113507623B (en) | 2021-07-09 | 2021-07-09 | Intelligent commodity selection system and method based on data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113507623A CN113507623A (en) | 2021-10-15 |
| CN113507623B true CN113507623B (en) | 2022-03-22 |
Family
ID=78012269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110774854.5A Active CN113507623B (en) | 2021-07-09 | 2021-07-09 | Intelligent commodity selection system and method based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113507623B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114240505B (en) * | 2021-12-21 | 2022-07-26 | 南京吐纳健康科技有限公司 | Big data based air purifier recommendation system and method |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611366A (en) * | 2015-10-26 | 2017-05-03 | 阿里巴巴集团控股有限公司 | Method and device for automatically selecting article |
| CN110020905A (en) * | 2018-01-09 | 2019-07-16 | 阿里巴巴集团控股有限公司 | A kind of method, apparatus and system of digitization selection |
| CN108960719B (en) * | 2018-06-29 | 2021-10-15 | 北京京东尚科信息技术有限公司 | Method and device for selecting products and computer readable storage medium |
| CN109493177A (en) * | 2018-11-12 | 2019-03-19 | 深圳云行智能科技有限公司 | A kind of intelligent Matching SKU selection method and its system and storage medium |
| CN112035735A (en) * | 2020-07-28 | 2020-12-04 | 康艾艺 | Short video-based commodity recommendation method, device and system |
| CN112232907A (en) * | 2020-09-29 | 2021-01-15 | 平潭综合实验区澄心贸易有限公司 | Commodity screening system for live broadcast sales and working method thereof |
| CN113946748A (en) * | 2020-10-13 | 2022-01-18 | 王琰 | Big data-based intelligent tourism target matching system |
| CN112967109A (en) * | 2021-03-15 | 2021-06-15 | 杭州邻汇网络科技有限公司 | Intelligent flash shop product selection method |
-
2021
- 2021-07-09 CN CN202110774854.5A patent/CN113507623B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113507623A (en) | 2021-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108322788B (en) | Advertisement display method and device in live video | |
| CN107798560A (en) | A kind of retail shop's individual character advertisement intelligent method for pushing and system | |
| CN112102037B (en) | Live E-commerce platform commodity content intelligent pushing management system based on big data | |
| CN106791901B (en) | Live broadcast channel ordering method and device | |
| US20200118168A1 (en) | Advertising method, device and system, and computer-readable storage medium | |
| CN104021162B (en) | A kind of method and device given a mark for multimedia resource | |
| CN113507623B (en) | Intelligent commodity selection system and method based on data analysis | |
| CN110881134B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN102957950A (en) | User implicit rating method for recommending video | |
| US20180225377A1 (en) | Method, server and terminal for acquiring information and method and apparatus for constructing database | |
| CN109409635B (en) | Agricultural product sensory data processing and acquiring method, device and system | |
| CN115983952A (en) | Custom garment design recommendation system | |
| CN108401186A (en) | A kind of smart television display control method based on recognition of face | |
| CN110889043A (en) | Diet recommendation system and method based on big data | |
| CN114219558A (en) | Intelligent agricultural product recommendation system based on data mining | |
| CN114005077A (en) | Audience evaluation data driven silent product video creation auxiliary method and device | |
| CN115225922A (en) | Method and device for delivering advertisement on live broadcast platform and electronic equipment | |
| CN110189179A (en) | A kind of precision advertisement broadcast method | |
| JP6677667B2 (en) | Data connection device, data connection method and program | |
| CN112801699A (en) | Advertisement putting method and device for advertisement platform | |
| CN115880037A (en) | Commodity recommendation method based on multi-project planning integration analysis | |
| Simons et al. | Mapping the preferences of adult and child consumers for California-grown navel oranges | |
| US20160357708A1 (en) | Data analysis method, data analysis apparatus, and recording medium having recorded program | |
| CN108399550B (en) | User grouping method | |
| CN112529048A (en) | Product display video aided design method and device based on perception experience |
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 |