CN117714722A

CN117714722A - Data analysis method and system for live shopping of electronic commerce

Info

Publication number: CN117714722A
Application number: CN202311660509.4A
Authority: CN
Inventors: 项玉凤; 王文
Original assignee: Yiming International Cultural Media Beijing Co ltd
Current assignee: Yiming International Cultural Media Beijing Co ltd
Priority date: 2023-12-05
Filing date: 2023-12-05
Publication date: 2024-03-15

Abstract

The invention discloses a data analysis method and a system for live shopping of an electronic commerce, wherein the method comprises the following steps: receiving a target anchor verification request and initial bonus sent by a willingness seller, comparing the target anchor verification request and initial bonus with a pre-stored blacklist, carrying out potential risk assessment on the target anchor according to live broadcast record data, establishing a relationship map on the target anchor of a potential risk label, judging whether abnormality exists according to the relationship map, establishing a live broadcast flow authenticity monitoring model, respectively establishing a first judgment standard and a second judgment standard for the potential risk label and the dangerous label anchor to carry out live broadcast monitoring, adding the target anchor into the blacklist if false flow is monitored, and sending a warning report to the willingness seller. The invention realizes the efficient and accurate identification and response of false flow, and improves the credibility and user satisfaction of the live shopping platform.

Description

Data analysis method and system for live shopping of electronic commerce

[ field of technology ]

The invention relates to the technical field of live broadcast of electronic commerce, in particular to a data analysis method and system for live broadcast shopping of electronic commerce.

[ background Art ]

The rapid growth in the field of electronic commerce has changed the traditional way of selling and marketing goods. With the advent of short video and live platforms, more and more manufacturers and sellers choose to promote and sell their products with the strength of the internet and social media. Wherein, it is a common mode to cooperate with live broadcast principals and with bloggers, which attract a great deal of attention and purchase of spectators by showing and popularizing various commodities to spectators in network live broadcast.

However, some dishonest broadcasters take false measures, such as using teams, purchasing attentors or using automated scripts to increase the number of live viewers and buyers to make false attentions and purchasing trendles in the live process, misguide sellers to think that there are a lot of real purchasing behavior, falsify purchasing transactions, make false purchasing records, and thus present false sales data to factories and sellers to obtain higher commissions or cooperation fees, whereas after live broadcast, false viewers initiate large-scale return requests to cancel the previous purchasing behavior, resulting in sellers not only losing sales, but also needing to bear return costs, seriously impairing the experience interests of sellers, and also compromising reputation and reputation of sellers, reducing competitiveness on electronic commerce platforms.

[ invention ]

In view of the above, the embodiment of the invention provides a data analysis method and a system for live shopping of an electronic commerce.

In a first aspect, an embodiment of the present invention provides a method for analyzing data of live shopping by an electronic commerce, where the method includes:

receiving a target anchor verification request and initial bonus sent by a willing seller, comparing the target anchor verification request and the initial bonus with a pre-stored blacklist, and if the target anchor verification request and the initial bonus are not matched with the pre-stored blacklist, acquiring live broadcast record data of other sellers for responding to the request and sharing the target anchor;

performing potential risk assessment on the target anchor according to the live broadcast record data, if no potential risk exists in the assessment, performing public opinion monitoring on the target anchor for a latest preset time period, if negative public opinion related to the target anchor is monitored, giving the target anchor a potential risk label, otherwise, giving the target anchor a potential security label; if the potential risk is estimated to exist, the target anchor is endowed with a potential risk label, and a seller sharing the live broadcast record data of the target anchor is paid with a prize of a preset proportion of initial prizes according to a preset intelligent contract;

establishing a relationship graph for a target anchor of the potential risk tag, judging whether the potential risk tag is abnormal according to the relationship graph, if so, replacing the potential risk tag of the target anchor with the risk tag, and sending a judgment result to a willingness seller;

And establishing a live broadcast flow authenticity monitoring model, respectively establishing a first judgment standard and a second judgment standard for the potential risk label and the risk label anchor for live broadcast monitoring, adding the target anchor into a blacklist if false flow is monitored, and sending a warning report to the willingness seller.

Aspects and any one of the possible implementations as described above, further providing an implementation, the method further including:

establishing a bonus pool which is bound with a seller account and receives the initial bonus, wherein the bonus pool is provided with read-only permission and only allows the authenticated seller to check the balance and transaction record of the bonus pool;

comparing the identity information of the target anchor with a pre-stored blacklist;

if the identity information of the target anchor does not match any record in the blacklist, pre-storing the initial prize into a prize pool, and initiating response requests of live broadcast record data of the target anchor to other sellers;

if the identity information of the target anchor matches the record in the blacklist, the initial prize is deducted from the system service point of the first preset proportion, the residual prize of the willingness seller is returned, and blacklist data of the target anchor is sent.

Aspects and any one of the possible implementations as described above, further providing an implementation in which the live recording data includes audience interaction data, product sales data, audience quality data, and interaction difference data between buyers and non-buyers;

The potential risk assessment for the target anchor according to the live broadcast record data specifically comprises the following steps:

calculating an average risk potential assessment valueWherein the mean risk potential evaluation value +.>The calculation formula of (2) is as follows:

wherein N represents the number of live record data, R _i A risk potential evaluation value representing the ith live record data;

wherein R represents a risk potential evaluation value, I _A Index representing sales data of a product, I _B Index for indicating audience interaction record, I _C Index indicating audience quality, I _D Index indicating difference between buyer and non-buyer interaction, w _A 、w _B 、w _C And w _D Respectively representing weights;

I _B ＝α _B ·C+β _B ·L+γ _B ·S，

wherein M is ^* Representing the total number of products purchased by a viewer on a live broadcast, M ₀ The number of repeated purchases of audience members is represented by Q, the sales of products is represented by Z, the return rate is represented by C, the comment number during live broadcast is represented by L, the praise number during live broadcast is represented by L, the share number during live broadcast is represented by n, the number of audience members is represented by x _i The longitude or latitude representing the ith viewer coordinate,mean value of longitude or latitude representing audience coordinates, md represents median of audience interaction, M ₁ Represent the number of interactions of the purchaser, M ₂ Representing the number of interactions of non-purchasers, alpha _A 、β _A 、α _B 、β _B 、γ _B 、α _C And beta _C Respectively represent the adjustment factors;

Comparing average risk potential assessment valuesAnd a preset risk threshold R ₀ When the average risk potential evaluation value +.>Greater than a preset risk threshold R ₀ And when the target anchor is identified as having potential risks.

In the aspect and any possible implementation manner described above, there is further provided an implementation manner, wherein the public opinion monitoring is performed on the target anchor for a last preset period of time, if negative public opinion related to the target anchor is detected, the target anchor is given a potential risk tag, otherwise, the target anchor is given a potential security tag, which specifically includes:

collecting names and keywords of a target anchor, and setting a monitoring time window;

searching in real time on news media and social media by using names and keywords of the target anchor, and acquiring public opinion related to the target anchor in a monitoring time window, wherein the public opinion comprises news articles, social media posts, comments and discussions;

emotion analysis is carried out on the public opinion content, and the public opinion content is classified into three emotion polarities of positive, negative or neutral,

the emotion analysis method comprises the following steps: converting the obtained public opinion into text data and preprocessing, splitting the text into words or sequences of words, using a predefined emotion vocabulary, distributing emotion scores to each word in the text, increasing scores for positive words, decreasing scores for negative words, and then calculating summary scores to classify emotion polarities of the text;

If the public opinion is marked as negative emotion and is directly related to the target anchor, recording the public opinion as negative public opinion;

carrying out content analysis on the negative public opinion, determining whether the negative public opinion is related to the behavior, the product or the language of a target anchor, and if so, counting the public opinion content into risk assessment;

counting the number of negative public opinion in a preset time period, and judging that potential risks exist if the number exceeds a preset number threshold;

the target anchor is given a potential risk tag, otherwise the target anchor is given a potential security tag.

Aspects and any possible implementation manner as described above, further provide an implementation manner, where the smart contract presetting method is as follows:

deploying an intelligent contract to enable all sellers to participate;

generating a unique contract address, said address being used for interaction;

the intelligent contract setting execution conditions are as follows: if the potential risk is not estimated, the initial prize is deducted from a system service point of a second preset proportion, and the residual prize of the willingness seller is returned;

if the potential risk exists, paying out the prize money with the preset proportion to the initial prize money to the seller sharing the target anchor live broadcast record data, and deducting the system service point with the third preset proportion from the initial prize money.

In the aspect and any possible implementation manner as described above, there is further provided an implementation manner, where the third preset proportion is greater than the second preset proportion, and the second preset proportion is greater than the first preset proportion.

The foregoing aspect and any possible implementation manner further provide an implementation manner, where the establishing a relationship graph for the target anchor of the potential risk tag, determining whether there is an abnormality according to the relationship graph, if yes, replacing the potential risk tag of the target anchor with a risk tag, and sending a determination result to the willingness seller, where the determining includes:

collecting fan concern data of a target anchor, and acquiring concern relations between the target anchor and the fan and between the fan and the fan;

according to the concern relation, connecting the target anchor as a root node and the fan as child nodes according to a tree structure to obtain a relation map represented by the tree structure;

detecting whether the relationship map is abnormal by using an abnormality detection algorithm, and/or,

collecting the data of the relationship atlas of the anchor for training, marking and classifying the relationship atlas as normal or abnormal, preprocessing the relationship atlas, converting the relationship atlas into an adjacent matrix or a node characteristic matrix, constructing a convolution neural network model for abnormal judgment, training the convolution neural network model through tag training data, and performing deep analysis on the relationship atlas of the anchor by using the trained convolution neural network after evaluation;

If the target anchor is judged to be abnormal, the potential risk label of the target anchor is replaced by the risk label, and the judgment result is sent to the willing seller.

Aspects and any one of the possible implementations as set forth above, further provide an implementation that builds a live traffic authenticity monitoring model, having a method comprising:

collecting live broadcast data, cleaning and preprocessing, and processing missing values and abnormal values;

constructing characteristic variables of an aggregation model, and calculating an aggregation degree G, wherein the calculation formula of the aggregation degree G is as follows:

A ₁ ＝N ₁ ·T ₁ ，

A ₂ ＝H ₂ ·ΔP ₂ ，

wherein A is ₁ Representing the attraction of the anchor, A ₂ Representing the attractive force of the product, D ₁ Indicating audience dispersion, F ₁ Represent the number of spectator interactions per minute S ₁ Representing the number of new viewers entered per minute, S ₂ Represents the number of product orders per minute, w ₁ And w ₂ Respectively representing the adjustment coefficients; n (N) ₁ Representing the number of vermicelli on the anchor, T ₁ Represents the accumulated live broadcast duration of the anchor, H ₂ Weighted integrated value, Δp, representing browsing amount, collection amount, and purchase amount of product ₂ Representing preferential price of the product, K _i Indicating the proportion of the audience in the ith area to the total audience, and n indicates the number of areas;

when live broadcast is carried out to a preset time T ₁ For a first preset time period delta T ₂ Starting the first live broadcast flow authenticity monitoring, and constructing a first preset time period delta T ₂ The aggregation degree G growth rate comparison model calculates a first comparison average growth rate R ₁ Calculating a first preset time period delta T ₂ True average growth rate R of aggregation G ₂ Comparing the true average growth rate R ₂ Average growth rate R in comparison with the first ₁ Wherein the first preset time period deltat ₂ The aggregation degree G growth rate comparison model is as follows:

wherein alpha is ₁ And b ₁ Representing a set growth rate curve parameter，t ₁ Representing a first preset time period DeltaT ₂ Time of (2);

during a first preset time period delta T ₂ After that, when it is detected that the aggregation level G is continuously decreased for a preset period of time DeltaT ₃ For a second preset time period delta T ₄ Starting the second live broadcast flow authenticity monitoring, and constructing a second preset time period delta T ₄ The second contrast average reduction rate R is calculated by using a contrast model of the reduction rate of the aggregation degree G ₃ Calculating a second preset time period delta T ₂ True average reduction rate R of aggregation degree G ₄ Comparing the true average reduction rate R ₄ Average reduction rate R from the second contrast ₃ Wherein the second preset time period deltat ₄ The comparative model of the aggregation degree G reduction rate is as follows:

wherein alpha is ₂ And b ₂ Indicating the set reduction rate curve parameter, _t2 representing a second preset time period DeltaT ₄ Time of (2);

when R is ₂ 〉R ₁ And R is ₄ 〈R ₃ And when the monitored live traffic is considered to be false traffic.

In the aspect and any possible implementation manner described above, there is further provided an implementation manner, where the establishing a first judgment standard and a second judgment standard for the risk potential tag and the risk potential tag anchor respectively for live broadcast monitoring specifically includes:

establishing a first judgment standard for a potential risk label anchor, wherein the first judgment standard adopts a first preset time period delta T ₂ The aggregation degree G growth rate comparison model is as follows:a second preset time period delta T ₄ The comparative model of the aggregation degree G reduction rate is as follows: />

Establishing a second judgment standard for the dangerous tag anchor, wherein the second judgment standard adopts adjustmentA first preset time period delta T ₂ The aggregation degree G growth rate comparison model is as follows:a second preset time period delta T ₄ The comparative model of the aggregation degree G reduction rate is as follows: />

Setting alpha ₁ >α ₁ ^* ，b ₁ >b ₁ ^* ，α ₂ <α ₂ ^* ，b ₂ >b ₂ ^* 。

In a second aspect, an embodiment of the present invention provides a data analysis system for live shopping by an electronic commerce using the data analysis method for live shopping by an electronic commerce, where the system includes:

the request management module is used for receiving the target anchor verification request and the initial prize sent by the willing seller, comparing the target anchor verification request with a pre-stored blacklist, and if the target anchor verification request and the initial prize are not matched with the pre-stored blacklist, acquiring live broadcast record data of the target anchor shared by other sellers in response to the request;

The risk assessment module is used for carrying out potential risk assessment on the target anchor according to the live broadcast record data, if the potential risk does not exist in the assessment, carrying out public opinion monitoring on the target anchor for a latest preset time period, if negative public opinion related to the target anchor is monitored, giving the target anchor a potential risk label, otherwise giving the target anchor a potential safety label; if the potential risk is estimated to exist, the target anchor is endowed with a potential risk label, and a seller sharing the live broadcast record data of the target anchor is paid with a prize of a preset proportion of initial prizes according to a preset intelligent contract;

the anomaly monitoring module is used for establishing a relationship graph of a target anchor of the potential risk tag, carrying out deep analysis on the relationship graph of the anchor by utilizing the convolutional neural network so as to detect whether anomalies exist, if so, replacing the potential risk tag of the target anchor with a dangerous tag, and sending a detection result to a willingness seller;

and the live broadcast monitoring module is used for establishing a live broadcast flow authenticity monitoring model, respectively establishing a first judgment standard and a second judgment standard for the potential risk label and the risk label anchor to carry out live broadcast monitoring, adding the target anchor into a blacklist if false flow is monitored, and sending a warning report to the willingness seller.

One of the above technical solutions has the following beneficial effects:

the method of the embodiment of the invention provides a data analysis method for live shopping of the electronic commerce, and has remarkable technical effects in the aspects of improving the risk prevention capability of a live platform, improving user experience, improving the credibility of the platform and the like.

[ description of the drawings ]

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for analyzing data of live shopping of an electronic commerce according to an embodiment of the present invention;

fig. 2 is a functional block diagram of a data analysis system for live shopping of an electronic commerce vendor according to an embodiment of the present invention.

[ detailed description ] of the invention

In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Fig. 1 is a flow chart of a data analysis method for live shopping of an electronic commerce according to an embodiment of the present invention. As shown in fig. 1, the method includes:

According to the data analysis method for live shopping of the electronic commerce, disclosed by the invention, by comprehensively analyzing and evaluating the potential risk of live record data of a target anchor, and combining with public opinion monitoring, false flow can be accurately identified, and negative public opinion is detected by utilizing public opinion data, so that misleading caused by unreal information is avoided; the multi-dimensional evaluation system is adopted, and comprises potential risk labels, relationship map anomaly judgment, live broadcast flow authenticity monitoring models and the like, and the reputation of the anchor is evaluated from different angles, so that the accuracy and the comprehensiveness are improved; by means of intelligent contract technology, corresponding operations including bonus payment, label replacement, blacklist addition and the like are automatically executed according to the evaluation result, and an intelligent risk handling strategy is realized; when the false flow is monitored, not only the target anchor is added into the blacklist, but also a warning report can be timely sent to the willingness seller, so that the benefit of the willingness seller is ensured, and the credibility of the platform and the satisfaction of the user are improved; by establishing a relationship graph, the association between the anchor is analyzed, so that whether abnormal behaviors exist can be more comprehensively judged, and the monitoring capability of false flow is enhanced; and establishing a live traffic authenticity monitoring model, and monitoring false traffic, so that an alarm can be sent to a willingness seller.

Therefore, the method for analyzing the live shopping data of the electronic commerce has remarkable technical effects in the aspects of improving the risk prevention capability of the live platform, improving the user experience, improving the credibility of the platform and the like.

In a preferred embodiment of the present invention, the method of the present invention further comprises:

The invention ensures the safety of the bonus pool by binding with the account of the seller and setting the read-only permission, and only the seller with identity verification can check the balance and transaction record of the bonus pool, thereby ensuring the safe storage and circulation of the bonus; the identity information of the target anchor is compared with the pre-stored blacklist, so that intelligent identity verification is realized, and when the identity information of the target anchor is matched with the record in the blacklist, the system can automatically process refund and send blacklist data, and the safety and stability of the platform are improved; when the target anchor identity information is not matched with the blacklist, pre-storing the initial prize into a prize pool, and initiating response requests to other sellers, the mechanism ensures the safe storage of the initial prize of the sellers, effectively avoids the transaction with potential risk anchors, and improves the safety of the platform; in the target anchor identity verification process, the system can inform relevant sellers in real time, the timeliness of information transmission is guaranteed, meanwhile, for the condition of matching with a blacklist, blacklist data of the target anchor are timely sent, and the quick response capability of the platform to potential risks is improved.

In a preferred embodiment of the present invention, the live recording data includes audience interaction data, product sales data, audience quality data, and interaction difference data between buyers and non-buyers;

I _B ＝α _B ·C+β _B ·L+γ _B ·S，

wherein M is ^* Representing the total number of products purchased by a viewer on a live broadcast, M ₀ The number of repeated purchases of audience members is represented by Q, the sales of products is represented by Z, the return rate is represented by C, the comment number during live broadcast is represented by L, the praise number during live broadcast is represented by L, the share number during live broadcast is represented by n, the number of audience members is represented by x _i The longitude or latitude representing the ith viewer coordinate, Mean value of longitude or latitude representing audience coordinates, md represents median of audience interaction, M ₁ Represent the number of interactions of the purchaser, M ₂ Representing the number of interactions of non-purchasers, alpha _A 、β _A 、α _B 、β _B 、γ _B 、α _C And beta _C Respectively represent the adjustment factors;

The method for calculating Md is as follows: and collecting audience interaction data, including comment quantity, praise quantity and sharing quantity, storing the interaction times of all audiences into a list, and sorting the list of the interaction times in ascending order or descending order so as to find a middle value, wherein if the length of the list is odd, the middle value is the value of the middle position of the list, and if the length of the list is even, the middle value is the average value of the middle two values. It should be noted that, the median of the viewer interaction is used to represent the overall interaction frequency of the viewer, and the median is generally more robust than the average, so that the influence of the extreme value on the result can be avoided.

The invention can judge the potential risk of the target anchor in real time by comparing the average potential risk evaluation value with the preset risk threshold value, thereby guaranteeing the timely monitoring and processing of the anchor risk, enabling the platform to rapidly respond and ensuring the safety and stability of the transaction.

In a preferred embodiment of the present invention, the present invention monitors public opinion of a target anchor for a latest preset period of time, if negative public opinion related to the target anchor is monitored, the target anchor is given a potential risk tag, otherwise the target anchor is given a potential security tag, and specifically includes:

It should be noted that, the content analysis of the negative public opinion may be performed by an existing method, or may be performed by the following method: firstly, text data including news articles, social media posts, comments and the like of negative public opinion needs to be collected, and the data can be acquired through web crawlers, API queries or data providers; then, cleaning and preprocessing the collected text data, including removing special characters, punctuation marks, HTML labels, stop words and the like; splitting the text into a sequence of words or terms; identifying and extracting keywords or phrases related to the target anchor in the text by a keyword extraction technology, wherein the keywords can comprise the name of the target anchor, the product name, the live broadcast activity name and the like; analyzing the text content using text mining and natural language processing techniques to determine if information related to the target anchor exists; assigning a relevance score to each text record indicating its degree of relevance to the target anchor, which may be a binary classification or score based; the relevance scores of the text records are output as relevant or irrelevant and recorded for further analysis or visualization.

According to the method, a plurality of information sources including news articles, social media posts, comments and discussions are covered by searching the news media and public opinion related to a target anchor on the social media in real time, and the information comprehensiveness and accuracy are ensured by multi-source public opinion collection; the emotion analysis method is adopted to classify the emotion polarity of the public opinion, the text data is preprocessed, a predefined emotion vocabulary is used to allocate emotion scores to each word in the text, the emotion polarity of the public opinion can be accurately judged by the emotion analysis method based on the vocabulary, and the analysis accuracy is improved; when determining whether the public opinion is related to the target anchor, carrying out content analysis to judge whether the public opinion content is directly related to the anchor behavior, product or language; the relevance judgment ensures that only negative public opinion related to the anchor is counted into risk assessment, and avoids the influence of irrelevant factors on the assessment result; and setting a preset number threshold, and judging that potential risks exist in the anchor when the number of negative public opinion exceeds the threshold in a preset time period. Such thresholding provides a clear criterion for the assessment, making the assessment more quantitative and operable; by adopting the real-time searching and monitoring method, the related information can be obtained rapidly after public opinion is generated, the timely discovery and processing of potential risks of a host are guaranteed, and the safety of platform transaction is guaranteed. The public opinion monitoring method combines the characteristics of multi-source public opinion acquisition, emotion analysis, relevance judgment and quantity threshold setting, has the characteristics of high accuracy, real-time performance and timeliness, provides reliable data support for potential risk assessment of a target anchor, and ensures the safety of a transaction platform.

In a preferred embodiment of the present invention, the smart contract presetting method of the present invention is as follows:

deploying an intelligent contract to enable all sellers to participate;

generating a unique contract address, said address being used for interaction;

In a preferred embodiment of the present invention, the third preset ratio is greater than the second preset ratio, which is greater than the first preset ratio.

In a preferred embodiment of the present invention, a relationship graph is established for a target anchor of a potential risk tag, whether an abnormality exists is determined according to the relationship graph, if the abnormality exists, the potential risk tag of the target anchor is replaced with a risk tag, and a determination result is sent to a willing seller, and the method specifically includes:

The invention collects the fan attention data of the target anchor, including the attention relation between the anchor and the fan to form a comprehensive data set, which ensures the sufficiency of the data source of the relation map and improves the accuracy of the relation analysis; based on the concern relation, a relation map of a tree structure is constructed, so that the structure of the relation map is clear, the subsequent analysis and judgment are convenient, and the representation mode of the tree structure is helpful for intuitively observing the association condition between the anchor and the vermicelli; the relationship map is subjected to deep analysis by using an anomaly detection algorithm and a convolutional neural network model, the anchor relationship map is trained and evaluated by adopting a deep learning technology, the precision and the accuracy of the relationship analysis are improved, and the relationship map anomaly judgment is more accurate by combining anomaly detection and deep analysis; once the relation map abnormality is detected, the potential risk label of the target anchor is replaced by the risk label, and the judgment result is timely sent to the willingness seller.

The method for analyzing the potential risk labels combines comprehensive data acquisition, tree structure relationship spectrum construction, anomaly detection and deep analysis, ensures the accuracy and the instantaneity of the relationship spectrum analysis, provides powerful support for potential risk assessment of a target anchor, and ensures the safety of a transaction platform.

In a preferred embodiment of the present invention, the present invention establishes a live traffic authenticity monitoring model having the steps of:

A ₁ ＝N ₁ ·T ₁ ，

A ₂ ＝H ₂ ·ΔP ₂ ，

wherein A is ₁ Representing the attraction of the anchor, A ₂ Representing the attractive force of the product, D ₁ Indicating audience dispersion, F ₁ Represent the number of spectator interactions per minute S ₁ Representing the number of new viewers entered per minute, S ₂ Represents the number of product orders per minute, w ₁ And w ₂ Respectively representing the adjustment coefficients; n (N) ₁ Representing the number of vermicelli on the anchor, T ₁ Represents the accumulated live broadcast duration of the anchor, H ₂ Weighted integrated value, Δp, representing browsing amount, collection amount, and purchase amount of product ₂ Representing the superiority of a productPrice is low, K _i Indicating the proportion of the audience in the ith area to the total audience, and n indicates the number of areas;

wherein alpha is ₁ And b ₁ Representing a set growth rate curve parameter, t ₁ Representing a first preset time period DeltaT ₂ Time of (2);

when R is ₂ 〉R ₁ And R is ₄ 〈R ₃ When the monitoring is carried out, the monitored live broadcast flow is determinedIs false traffic.

The live broadcast flow authenticity monitoring model has the technical effects of comprehensive data processing, comprehensive characteristic variable construction, dynamic monitoring model establishment, multi-parameter comprehensive judgment, real-time false flow judgment and the like, provides an efficient and accurate flow monitoring means for an e-commerce live broadcast shopping platform, and ensures the authenticity and safety of live broadcast transactions.

In a preferred embodiment of the present invention, the present invention establishes a first judgment standard and a second judgment standard for a potential risk tag and a risk tag anchor, respectively, to perform live broadcast monitoring, and specifically includes:

Establishing a second judgment standard for the dangerous tag anchor, wherein the second judgment standard adopts an adjusted first preset time period delta T ₂ The aggregation degree G growth rate comparison model is as follows:a second preset time period delta T ₄ The comparative model of the aggregation degree G reduction rate is as follows: />

The first judgment standard and the second judgment standard have the technical characteristics of pertinence, dynamic monitoring, parameter adjustment, real-time judgment and the like, and a high-efficiency and accurate solution is provided for live broadcast monitoring of potential risk tags and danger tag broadcasters.

On the basis of the above, please refer to fig. 2 in combination, an embodiment of the present invention provides a data analysis system for live shopping of an electronic commerce, the system includes:

According to the invention, the risk of the target anchor can be comprehensively and multi-angularly estimated through the risk estimation module by combining live broadcast record data and public opinion monitoring, and the effective identification of various potential risks is comprehensively ensured, so that a powerful foundation is provided for subsequent monitoring; the anomaly monitoring module utilizes the convolutional neural network to carry out depth analysis on the relationship map of the anchor, can efficiently detect whether the potential risk label anchor has anomalies, and the intelligent monitoring means ensures the timely discovery and processing of the anomalies and reduces the possibility of potential risks; the live broadcast monitoring module establishes a live broadcast flow authenticity monitoring model, and establishes corresponding judgment standards according to different label anchor. The dynamic monitoring mechanism enables the system to adjust the monitoring strategy according to actual conditions, and ensures accurate monitoring of false flow; once the false flow is monitored, the system can add the target anchor to the blacklist in time and send a warning report to the willingness seller, so that quick early warning and handling are realized, the propagation of the false flow is effectively restrained by the mechanism, and the fairness and the reality of live shopping are ensured.

The data analysis system provided by the invention has the technical characteristics of comprehensive risk assessment, intelligent anomaly monitoring, dynamic live broadcast monitoring, timely early warning and the like, provides reliable guarantee for live broadcast shopping of electronic commerce, and improves the safety and the trust of the whole live broadcast transaction system.

On the basis of the above, there is also provided a computer readable storage medium on which a computer program stored which, when run, implements the above method.

It should be appreciated that the systems and modules thereof shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may then be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application may be implemented not only with hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also with software, such as executed by various types of processors, and with a combination of the above hardware circuitry and software (e.g., firmware).

It should be noted that, the advantages that may be generated by different embodiments may be different, and in different embodiments, the advantages that may be generated may be any one or a combination of several of the above, or any other possible advantages that may be obtained.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations of the present application may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this application, and are therefore within the spirit and scope of the exemplary embodiments of this application.

Meanwhile, the present application uses specific words to describe embodiments of the present application. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present application may be combined as suitable.

Furthermore, those skilled in the art will appreciate that the various aspects of the invention are illustrated and described in the context of a number of patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.

The computer storage medium may contain a propagated data signal with the computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take on a variety of forms, including electro-magnetic, optical, etc., or any suitable combination thereof. A computer storage medium may be any computer readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer storage medium may be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or a combination of any of the foregoing.

The computer program code necessary for operation of portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, scala, smalltalk, eiffel, JADE, emerald, C ++, c#, vb net, python, etc., a conventional programming language such as C language, visual Basic, fortran 2003, perl, COBOL 2002, PHP, ABAP, dynamic programming languages such as Python, ruby and Groovy, or other programming languages, etc. The program code may execute entirely on the user's computer or as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or the use of services such as software as a service (SaaS) in a cloud computing environment.

Furthermore, the order in which the elements and sequences are presented, the use of numerical letters, or other designations are used in the application and are not intended to limit the order in which the processes and methods of the application are performed unless explicitly recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present application. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

Likewise, it should be noted that in order to simplify the presentation disclosed herein and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are presented in the claims are required for the subject application. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the numbers allow for adaptive variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., cited in this application is hereby incorporated by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the present application, documents that are currently or later attached to this application for which the broadest scope of the claims to the present application is limited. It is noted that the descriptions, definitions, and/or terms used in the subject matter of this application are subject to such descriptions, definitions, and/or terms if they are inconsistent or conflicting with such descriptions, definitions, and/or terms.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present application. Other variations are also possible within the scope of this application. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present application may be considered in keeping with the teachings of the present application. Accordingly, embodiments of the present application are not limited to only the embodiments explicitly described and depicted herein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. A data analysis method for live shopping of an electronic commerce, the method comprising:

2. The method for analyzing data of live shopping by an electronic commerce according to claim 1, wherein the method further comprises:

3. The data analysis method of live shopping by an e-commerce according to claim 2, wherein the live recording data includes audience interaction data, product sales data, audience quality data, and interaction difference data between a purchaser and a non-purchaser;

I _B ＝α _B ·C+β _B ·L+γ _B ·S，

comparing the average risk potential evaluation value R with a preset risk threshold value R ₀ When the average risk potential evaluation value R is greater than the preset risk threshold value R ₀ And when the target anchor is identified as having potential risks.

4. The method for analyzing data of live shopping by an electronic commerce merchant according to claim 1, wherein the monitoring of public opinion of the target anchor for a latest preset period of time, if negative public opinion related to the target anchor is monitored, the target anchor is given a potential risk tag, otherwise, the target anchor is given a potential security tag, specifically includes:

5. The data analysis method for live shopping by an e-commerce according to claim 4, wherein the intelligent contract presetting method comprises the following steps:

deploying an intelligent contract to enable all sellers to participate;

generating a unique contract address, said address being used for interaction;

6. The method for analyzing data of live shopping by an electronic commerce according to claim 5, wherein the third preset proportion is larger than the second preset proportion, and the second preset proportion is larger than the first preset proportion.

7. The method for analyzing data of live shopping by an electronic commerce according to claim 1, wherein the creating a relationship graph for the target anchor of the potential risk tag, judging whether there is an abnormality according to the relationship graph, if yes, replacing the potential risk tag of the target anchor with a risk tag, and transmitting the judgment result to the willingness seller, specifically comprises:

8. The method for analyzing data of live shopping by an electronic commerce according to claim 1, wherein the establishing a live traffic authenticity monitoring model includes:

A ₁ ＝N ₁ ·T ₁ ，

A ₂ ＝H ₂ ·ΔP ₂ ，

9. The method for analyzing data of live shopping by an electronic commerce according to claim 8, wherein the establishing a first judgment standard and a second judgment standard for the potential risk tag and the risk tag anchor respectively for live monitoring specifically comprises:

Setting alpha ₁ ＞α ₁ ^* ，b ₁ ＞b ₁ ^* ，α ₂ ＜α ₂ ^* ，b ₂ ＞b ₂ ^* 。

10. A data analysis system for live shopping by an electronic commerce using the data analysis method for live shopping by an electronic commerce according to any one of claims 1 to 9, the system comprising: