Image data analysis system for electronic commerce

Abstract

The invention discloses an image data analysis system for e-commerce, relates to the technical field of e-commerce, and discloses an image data analysis system for e-commerce, which comprises a data acquisition module, an image analysis module and an image modification module, through the data acquisition module, the click rate of commodity images of merchants and the sales volume of corresponding commodities can be collected, the image analysis module is set, and the image modification module is set. The click rate of the commodity image of the merchant can be analyzed, whether the commodity image of the merchant is an abnormal commodity image or not is judged, the merchant can adjust and modify the abnormal commodity image in time, the image modification module is arranged, the merchant can sort the commodity sales according to the influence of the commodity image on the commodity sales, and the commodity sales quality is improved. A merchant can firstly modify the commodity image of the commodity which is greatly influenced, so that the sales loss caused by the commodity image is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce, and more specifically, it relates to an image data analysis system for e-commerce. Background Art

[0002] E-commerce generally refers to a new business operation model in which extensive commercial trade activities are carried out around the world, in an open network environment of the Internet, based on the client / server application mode, and the buyer and seller conduct various business activities without meeting face to face, realizing consumers' online shopping, online transactions between merchants, online electronic payment, and various business activities, transaction activities, financial activities, and related integrated service activities. Governments, scholars, and business people in various countries have given many different definitions according to their own positions and the angles and degrees of participation in e-commerce. E-commerce is divided into: ABC, B2B, B2C, C2C, B2M, M2C, B2A (i.e., B2G), C2A (i.e., C2G), O2O, etc.

[0003] Currently, when conducting e-commerce shopping, merchants will place product images on the product page. However, the product images can also affect the click and sales of the products. If the image of a product fails to attract users to click, it will also affect the sales volume of the product. The current e-commerce image data analysis system does not consider the impact of product images on the image click volume, and at the same time, it is unable to distinguish which product images have a greater impact on the sales volume of the products. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an image data analysis system for e-commerce.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] An image data analysis system for e-commerce, including a data acquisition module, an image analysis module, and an image modification module;

[0007] The data acquisition module is used to acquire the image data of the merchant and send the image data to the server;

[0008] The image analysis module is used to analyze the click volume of the merchant's product images and determine whether the merchant's product images are abnormal product images;

[0009] The image modification module is used to analyze the abnormal product images of the merchant and sort the abnormal product images that need to be modified in order, specifically:

[0010] Obtain the click warning value Ks of each abnormal product image of the merchant in the thirty days before the current time of the system;

[0011] Obtain the daily sales volume of the product corresponding to each abnormal product image of the merchant in the thirty days before the current system time. Mark the product corresponding to each abnormal product image of the merchant as an abnormally sold product. Set a reasonable sales volume corresponding to the daily sales volume of each abnormally sold product. Compare the daily sales volume with the reasonable sales volume. When the daily sales volume is less than the reasonable sales volume, mark the daily sales volume of the abnormally sold product as a dim sales volume. Calculate the difference between the reasonable sales volume and the dim sales volume to obtain a reasonable sales difference. Sum up the reasonable sales differences of the abnormally sold product in the thirty days before the current system time to obtain a total reasonable sales difference, and mark it as Es; Sort the dim sales volumes according to the time sequence, calculate the time difference between the times of two adjacent sorted dim sales volumes to obtain an expected difference duration, sum up the expected difference durations to obtain a total expected difference duration, and mark it as Bs. Use the formula Obtain the dim sales value Fx, where n1 and n2 are both preset proportional coefficients;

[0012] Use the formula Gz = Ks×y1 + Fx×y2 to obtain the design value of each abnormal product image of the merchant, sort the design values of each abnormal product image of the merchant in descending order of value, and display the sorting result on the terminal corresponding to the merchant.

[0013] Further, the image data is the click volume of a single product image of the merchant and the sales volume of the corresponding product.

[0014] Further, the image analysis module is used to analyze the click volume of the product image of the merchant to determine whether the product image of the merchant is an abnormal product image. Specifically:

[0015] Mark the click volume of a single product image of the merchant every day in the thirty days before the current system time as the actual click volume. Set a supposed click volume corresponding to the actual click volume of a single product image of the merchant every day. Compare the actual click volume with the supposed click volume. When the actual click volume is lower than the supposed click volume, mark the actual click volume as an abnormal click volume. Calculate the difference between the supposed click volume and the abnormal click volume to obtain a click volume difference, and mark it as Ms, s = 1, 2,..., s;

[0016] Use the formula Calculate to obtain the abnormal click value Je. Sort the dates corresponding to the abnormal click values in chronological order. Calculate the time difference between the dates corresponding to two adjacent abnormal click values to obtain an abnormal time difference. Sum up all the abnormal time differences and take the average to obtain an average abnormal time difference and mark it as Hk;

[0017] Obtain the number of days with abnormal click volume in the thirty days before the current system time, and mark it as Ty.

[0018] Furthermore, use the formula Obtain the click warning value Kx of a single commodity image of a merchant; where m1, m2, and m3 are all preset proportional coefficients; set the click warning value threshold as Js. When the click warning value Kx of a single commodity image of a merchant is ≥ the click warning value threshold Js, mark the commodity image of this merchant as an abnormal commodity image. When the click warning value Kx of a single commodity image of a merchant < the click warning value threshold Js, then mark the commodity image of this merchant as a normal commodity image.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. Set up a data acquisition module to collect the click volume of the commodity images of merchants and the sales volume of the corresponding commodities. Set up an image analysis module to analyze the click volume of the commodity images of merchants and determine whether the commodity images of merchants are abnormal commodity images, enabling merchants to adjust and modify abnormal commodity images in a timely manner;

[0021] 2. Set up an image modification module, enabling merchants to sort according to the magnitude of the influence on commodity sales caused by the commodity images. Merchants can first modify the commodity images of the commodities with greater influence to reduce the sales volume loss caused by the commodity images. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is the principle block diagram of the present invention;

[0023] Figure 2 is the flow block diagram of the image analysis module of the present invention;

[0024] Figure 3 is the flow block diagram of the image modification module of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0025] Example 1

[0026] Refer to Figure 1 - Figure 2 , an image data analysis system for e-commerce, including a data acquisition module and an image analysis module;

[0027] The data acquisition module is used to collect the image data of merchants and send the image data to the server; the image data is the click volume of a single commodity image of a merchant and the sales volume of the corresponding commodity;

[0028] The image analysis module is used to analyze the click volume of the commodity images of merchants and determine whether the commodity images of merchants are abnormal commodity images. Specifically:

[0029] Mark the click volume of each single product image of the merchant every day within thirty days before the current system time as the actual click volume. Assume that each actual click volume of the single product image of the merchant every day corresponds to a supposed click volume. Compare the actual click volume with the supposed click volume. When the actual click volume is lower than the supposed click volume, mark the actual click volume as an abnormal click volume. Calculate the difference between the supposed click volume and the abnormal click volume to obtain the click volume difference, and mark it as Ms, where s = 1, 2, …, s;

[0030] Set the click volume difference coefficient as Kd, where d = 1, 2, 3, …, d; K1 < K2 < K3 < … < Kd. Set a range of click volume differences corresponding to each click volume difference coefficient, including (0, M1], (M1, M2], …, (Ms - 1, Ms]. When Ms ∈ (0, M1], the corresponding response coefficient value is K1;

[0031] Use the formula Calculate to obtain the abnormal click value Je. Sort the dates corresponding to the abnormal click values in chronological order. Calculate the time difference between the dates corresponding to two adjacent abnormal click values to obtain the abnormal time difference. Sum up all the abnormal time differences and take the average to obtain the average abnormal time difference and mark it as Hk;

[0032] Obtain the number of days with abnormal click volume within thirty days before the current system time, and mark it as Ty;

[0033] Use the formula Obtain the click warning value Kx of the merchant's single product image; where m1, m2, and m3 are all preset proportionality coefficients. Set the click warning value threshold as Js. When the click warning value Kx of the merchant's single product image ≥ the click warning value threshold Js, mark the product image of this merchant as an abnormal product image. When the click warning value Kx of the merchant's single product image < the click warning value threshold Js, then mark the product image of this merchant as a normal product image.

[0034] Embodiment 2

[0035] Refer to Figure 3 , on the basis of Embodiment 1, it further includes an image modification module. The image modification module is used to analyze the abnormal product images of the merchant and sort the abnormal product images that need to be modified in order, specifically:

[0036] Obtain the click warning value Ks of each abnormal product image of the merchant within thirty days before the current system time;

[0037] Obtain the daily sales volume of the product corresponding to each abnormal product image of the merchant in the thirty days before the current system time. Mark the product corresponding to each abnormal product image of the merchant as an abnormally sold product. Set a reasonable sales volume corresponding to the daily sales volume of each abnormally sold product. Compare the daily sales volume with the reasonable sales volume. When the daily sales volume is less than the reasonable sales volume, mark the daily sales volume of the abnormally sold product as a dim sales volume. Calculate the difference between the reasonable sales volume and the dim sales volume to obtain a reasonable sales difference. Sum up the reasonable sales differences of the abnormally sold product in the thirty days before the current system time to obtain a total reasonable sales difference, and mark it as Es. Sort the dim sales volumes according to time sequence, calculate the time difference between the times of two adjacent sorted dim sales volumes to obtain an expected difference duration, sum up the expected difference durations to obtain a total expected difference duration, and mark it as Bs. Use the formula Obtain the dim sales value Fx, where n1 and n2 are both preset proportional coefficients;

[0038] Use the formula Gz = Ks×y1 + Fx×y2 to obtain the design value of each abnormal product image of the merchant, sort the design values of each abnormal product image of the merchant in descending order of value, and display the sorting result on the terminal corresponding to the merchant.

[0039] Working principle:

[0040] Set up a data collection module to collect the click volume of the merchant's product images and the sales volume of the corresponding products. Set up an image analysis module to analyze the click volume of the merchant's product images and determine whether the merchant's product images are abnormal product images, so that the merchant can timely adjust and modify the abnormal product images. Set up an image modification module so that the merchant can sort according to the magnitude of the influence of the product sales caused by the product images. The merchant can first modify the product images of the products with greater influence to reduce the sales volume loss caused by the product images.

[0041] The above is only the preferred implementation mode of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and refinements should also be regarded as the protection scope of this template.

Claims

1. An image data analysis system for e-commerce, characterized in that, It includes a data acquisition module, an image analysis module, and an image modification module; The data acquisition module is used to collect the image data of merchants and send the image data into the server; The image analysis module is used to analyze the click volume of the product images of merchants and determine whether the product images of merchants are abnormal product images; The image modification module is used to analyze the abnormal product images of merchants, and sort the abnormal product images that need to be modified in order. Specifically: Obtain the click warning value Ks of each abnormal product image of the merchant in the thirty days before the current system time; Obtain the daily sales volume of the product corresponding to each abnormal product image of the merchant in the thirty days before the current system time, mark the product corresponding to each abnormal product image of the merchant as an abnormal sales product, set a reasonable sales volume corresponding to the daily sales volume of each abnormal sales product, compare the daily sales volume with the reasonable sales volume. When the daily sales volume is less than the reasonable sales volume, mark the daily sales volume of this abnormal sales product as a dim sales volume, calculate the difference between the reasonable sales volume and the dim sales volume to obtain a reasonable sales difference, sum up the reasonable sales differences of this abnormal sales product in the thirty days before the current system time to obtain a total reasonable sales difference, and mark it as Es; Sort the dull sales volumes in chronological order, calculate the time difference between the times of two adjacent dull sales volumes after sorting to obtain the expected difference duration, sum up the expected difference durations to obtain the total expected difference duration, and mark it as Bs. Use the formula to obtain the dull sales value Fx, where n1 and n2 are both preset proportionality coefficients; Use the formula Gz = Ks×y1 + Fx×y2 to obtain the design value of each abnormal product image of the merchant, sort the design values of each abnormal product image of the merchant in descending order of value, and display the sorting result on the terminal corresponding to the merchant.

2. The image data analysis system for e-commerce according to claim 1, wherein The image data is the click volume of a single product image of the merchant and the sales volume of the corresponding product.

3. An image data analysis system for e-commerce according to claim 2, wherein, The image analysis module is used to analyze the click volume of the product images of merchants and determine whether the product images of merchants are abnormal product images. Specifically: Mark the click volume of a single product image of the merchant every day in the thirty days before the current system time as the actual click volume, set a supposed click volume corresponding to the actual click volume of a single product image of the merchant every day, compare the actual click volume with the supposed click volume. When the actual click volume is lower than the supposed click volume, mark this actual click volume as an abnormal click volume, calculate the difference between the supposed click volume and the abnormal click volume to obtain a click volume difference, and mark it as Ms, s = 1, 2,..., s; Using the formula Calculate to obtain the abnormal click value Je, sort the dates corresponding to the abnormal click values in chronological order, calculate the time difference between the dates corresponding to two adjacent abnormal click values to obtain the abnormal time difference, sum up all the abnormal time differences and take the average to obtain the average abnormal time difference and mark it as Hk; Obtain the number of days with abnormal click volume in the thirty days before the current system time and mark it as Ty.

4. An image data analysis system for e-commerce according to claim 3, characterized in that, Using the formula Obtain the click alarm value Kx of a single product image of a merchant; where m1, m2, and m3 are all preset proportionality coefficients; set the click alarm value threshold as Js. When the click alarm value Kx of a single product image of a merchant is ≥ the click alarm value threshold Js, mark the product image of this merchant as an abnormal product image. When the click alarm value Kx of a single product image of a merchant is < the click alarm value threshold Js, then mark the product image of this merchant as a normal product image.

Images