E-commerce Data Research Articles

Implementing Big Data Analytics in E-Commerce: Vendor and Customer View

Due to the information revolution, one of the biggest challenges of e-commerce is the huge data that needs to be processed and analyzed to gain benefits from it. Big Data Analytics (BDA) aims to improve the decision-making process by analyzing and understanding big data, e.g., messages, social media posts, etc. Furthermore, BDA capabilities are used in e-commerce activities as a key growth direction to increase vendors' revenues and attract customers. Based on the importance of analyzing big data and its advantages to e-commerce operations, this paper aims to study the values of implementing BDA in e-commerce to both vendors and consumers. Fifteen papers are selected to analyze the impacts of analyzing big data in e-commerce. Electronic vendors (E-vendors) use BDA to gain the competitive advantages they need to understand consumer behaviour and increase their income by improving customer loyalty. Besides, recommendation systems derived from BDA personalize the searching and shopping experience of the customers. However, there are some negative effects derived from applying BDA in e-commerce, such as shopping addiction. In addition, e-vendors have to deal with expensive BDA tools and professionals. In conclusion, even though BDA enhances consumers' and vendors' electronic shopping experience, the rapid growth of data is still challenging.

Prediction of User Consumption Behavior Data Based on the Combined Model of TF-IDF and Logistic Regression

In the era of the rapid development of computers and the Internet, e-commerce has become a part of the economy of many countries. Therefore, how to use historical data of user consumption behavior to predict user shopping intentions accurately and subsequent personalized recommendations has turned into research hotspots in the field of e-commerce. This paper conducts a basic analysis of JD’s e-commerce data based on machine learning. Specifically, this article constructs user-product matrix and product and user clustering by means of text processing and clustering, as well as implements a logistic regression classifier to predict the user’s purchase intention of products in a certain target category in the next 5 days. Based on the JD competition data set, this article has a prediction accuracy rate of 98%. This can help e-commerce companies make better decisions.

A Knowledge-Enhanced Dialogue Model Based on Multi-Hop Information with Graph Attention

With the continuous improvement of the e-commerce ecosystem and the rapid growth of e-commerce data, in the context of the e-commerce ecosystem, consumers ask hundreds of millions of questions every day. In order to improve the timeliness of customer service responses, many systems have begun to use customer service robots to respond to consumer questions, but the current customer service robots tend to respond to specific questions. For many questions that lack background knowledge, they can generate only responses that are biased towards generality and repetitiveness. To better promote the understanding of dialogue and generate more meaningful responses, this paper introduces knowledge information into the research of question answering system by using a knowledge graph. The unique structured knowledge base of the knowledge graph is convenient for knowledge query, can acquire knowledge faster, and improves the background information needed for answering questions. To avoid the lack of information in the dialogue process, this paper proposes the Multi-hop Knowledge Information Enhanced Dialogue-Graph Attention (MKIED-GA) model. The model first retrieves the problem subgraph directly related to the input information from the entire knowledge base and then uses the graph neural network as the knowledge inference module on the subgraph to encode the subgraph. The graph attention mechanism is used to determine the one-hop and two-hop entities that are more relevant to the problem to achieve the aggregation of highly relevant neighbor information. This further enriches the semantic information to provide a better understanding of the meaning of the input question and generate appropriate response information. In the process of generating a response, a multi-attention flow mechanism is used to focus on different information to promote the generation of better responses. Experiments have proved that the model presented in this article can generate more meaningful responses than other models.

Management Model on Electronic Commerce Data Based on Cloud Computing

Judging from the current situation, China’s e‐commerce companies are developing more and more rapidly, and the sales records of e‐commerce platforms continue to break through. However, in this rapid development, many problems have been exposed. Because B2C e‐commerce business management methods are different from traditional business management methods, there are many fundamental differences between the risks faced by traditional e‐commerce companies and traditional enterprises. Compared with traditional enterprises, the risks faced by e‐commerce enterprises have greater uncertainty and complexity, and the related risk management measures have not fully kept up with the development speed of B2C e‐commerce and are still in the stage of continuous exploration. With the popularity of e‐commerce and higher applicability, the research on e‐commerce data management model is urgent. In order to study the e‐commerce data management model, theoretical research and commercial application of cloud computing are discussed in depth from literature analysis, comparative analysis, graphical analysis, and case analysis. I hope to expand the application research of cloud computing platform in e‐commerce. Finally, it proves that e‐commerce has a good development advantage in the future development.

Metode Alternatif dalam Pencarian Peringkat E-Commerce di Indonesia Berdasarkan Rating Pelanggan

The existence of the internet raises an online trading system using applications. The rise of online trading systems has triggered the emergence of various e-commerce in Indonesia that provide various kinds of customer needs. This also causes problems for customers, namely the difficulty in choosing quality e-commerce. The effort to overcome this problem is to rank e-commerce in Indonesia based on customer ratings. The method commonly used for ranking is the analytical hierarchy process (AHP) method, but in practice there are several variables that are not found in e-commerce so the AHP method cannot be used. The alternative method chosen is the ant colony optimization (ACO) method. The feasibility test of the ACO method in searching rankings for e-commerce data needs to be done because not all variables are in e-commerce. Simulations for ranking search are carried out using 2 generated data scenario with analytical hierarchy process (AHP) and ant colony optimization (ACO) method. The simulation results show that the ACO method is feasible to be used for ranking with blank data, then the application of the ACO method for e-commerce data in Indonesia. The best taboo results are based on the highest opportunity value and the highest correlation coefficient, namely in the second taboo, with three major ratings, namely JD, SP, and TP

Research on the Influence of E-commerce Entrepreneurship Education on the Development of Marine Economy Based on the Entropy Method

Qin, H., 2020. Research on the influence of e-commerce entrepreneurship education on the development of marine economy based on the entropy method. In: Liu, X. and Zhao, L. (eds.), Today's Modern Coastal Society: Technical and Sociological Aspects of Coastal Research. Journal of Coastal Research, Special Issue No. 111, pp. 263–267. Coconut Creek (Florida), ISSN 0749-0208.With the development of ocean cultivation, various projects related to the sea have been implemented, and various marine activities have been increasing. The growth rate and scale of the marine economy are constantly improving, and the change of economic structure will inevitably have an impact on the hierarchical structure of education. This paper designs a service framework for entrepreneurship education based on e-commerce and big data. The entropy method is used to objectively assign weights, and the index value is added up to determine the intensity of entrepreneurship education. In terms of educational support, the stochastic frontier method (SFA) is used to select educational indicators related to the marine industry and analyze them from the perspective of e-commerce entrepreneurship education. During the simulation experiment, the accuracy of the algorithm and the coupling of indicators are analyzed, the data obtained through the information platform are verified by calculation, and their effects on the weight of education-related indicators is presented.

Achieving Secure Search over Encrypted Data for e-Commerce

The advances of Internet technology has resulted in the rapid and pervasive development of e-commerce, which has not only changed the production and operation mode of many enterprises, but also affected the economic development mode of the whole society. This trend has incurred a strong need to store and process large amounts of sensitive data. The traditional data storage and search solutions cannot meet such requirements. To tackle this problem, in this article, we proposed Consortium Blockchain-based Distributed Secure Search (CBDSS) Scheme over encrypted data in e-Commerce environment. By integrating the blockchain and searchable encryption model, sensitive data can be effectively protected. The consortium blockchain can ensure that only authorized nodes can join the system. To fairly assign nodes for the search tasks, we developed an endorsement strategy in which two agent roles are set up to divide and match the search tasks with the virtual resources according to the load capacity of each node. The security analysis and experiments are conducted to evaluate the performance of our proposed scheme. The evaluation results have proved the reliability and security of our scheme over existing methods.

VISUALISASI DATA E-DAGANG MENGGUNAKAN AUGMENTASI REALITI

Data visualization is viewed as a significant element in data analysis and communication. As the data engagement becomes more and more complex, visual presentation of data does help users understand the data. So far, two-dimensional (2D) data visuals are often used for the data visualization process, but the lack of depth dimension leads to inefficient and limited understanding of the data. Therefore, the effectiveness of augmented reality (AR) in data visualization was studied through the development of an AR Data Visualization application using E-commerce data. Machine learning models are also involved in the development of this AR application for the provision of data using predictive analysis functions. To provide quality E-commerce data and an optimal machine learning model, the data science process is carried out using the python programming language. The E-commerce data selected for this study is open data taken through the Kaggle Website. This database has 9994 data numbers and 21 attributes. This AR data visualization application will make it easier for users to understand the E-commerce data in-depth through the use of AR technology and be able to visualize the forecasts for sales profit based on the algorithm model "Auto-Regressive Integrated Moving Average" (ARIMA).

THE PRECONDITIONS FOR THE EMERGENCE AND DRIVERS OF MARKETPLACES DEVELOPMENT IN E-COMMERCE

The virtual environment is getting closer to the person. From an information field designed for information exchange, it is transformed into a full-fledged business environment. A significant number of business processes today are conducted via the Internet, which necessitates an alternative approach to the organization of key aspects of management. The rapid development of technology leads to the emergence of novel forms of interaction between contractors, increases their range, involves not only manufacturers and buyers but also other companies, including logistics, banking, brokerage, into the processes, which leads to increased overall business activity. Intensification of business interaction leads to the emergence of new business models. In particular, the current stage is characterized by the transition from online stores to marketplaces, which are full-fledged trading platforms where there is supply and demand. The article examines the prerequisites for the emergence of marketplaces as one of the tools of digital marketing and provides a brief description of them. Thus, the analysis of the level of Internet penetration in the world and in Ukraine confirmed the course of digitalization of society. The study of analytical reports made it possible to confirm the rapid spread of access to the network and the involvement in the virtual environment of an increasing number of users in different age groups and from different regions. Such penetration allows to take into account globalization, eliminate barriers to access, and ensure greater involvement of counter-parties in market processes taking place in the network. The analysis of e-commerce indicators has confirmed the assumption that the intensity of use of e-commerce services correlates with the level of the Internet penetration. Joint analysis of the level of the Internet penetration and e-commerce data has revealed the key characteristics of the modern user of the network as a buyer. In particular, to outline his or her behavioral and consumer preferences, note the attitude to advertising, the desired level of service and the need to ensure data confidentiality. Drawing up a portrait of the consumer allows to understand his or her network activity and take it into account in the formation of communication, sales organization and product promotion. Tracking consumer activity also permits making a modeling impact on it, pushing the buyer to certain actions from information to purchase. Assessment of the current situation in Ukraine allows to predict a corresponding increase in activity in the field of electronic sales. The various business models used in e-commerce are considered, and their functionality is depicted. The key factors that led to the emergence of marketplaces are highlighted and the impact of each of them is considered separately. The classification of marketplaces according to different classification features is given and the advantages of use are outlined. The tendency to emergence of new formats of marketplaces and expansion of their functionality is revealed, which significantly increases the possibilities of users from the position of organization of work and ensuring circulation of information flows. The main drivers of formation of marketplaces in the world and Ukraine are singled out. The vector of their development is traced.

Pre-processing of E-Commerce Data for Customer Churn Prediction using Data Mining

Pre-processing of E-Commerce Data for Customer Churn Prediction using Data Mining

Shopper intent prediction from clickstream e-commerce data with minimal browsing information

We address the problem of user intent prediction from clickstream data of an e-commerce website via two conceptually different approaches: a hand-crafted feature-based classification and a deep learning-based classification. In both approaches, we deliberately coarse-grain a new clickstream proprietary dataset to produce symbolic trajectories with minimal information. Then, we tackle the problem of trajectory classification of arbitrary length and ultimately, early prediction of limited-length trajectories, both for balanced and unbalanced datasets. Our analysis shows that k-gram statistics with visibility graph motifs produce fast and accurate classifications, highlighting that purchase prediction is reliable even for extremely short observation windows. In the deep learning case, we benchmarked previous state-of-the-art (SOTA) models on the new dataset, and improved classification accuracy over SOTA performances with our proposed LSTM architecture. We conclude with an in-depth error analysis and a careful evaluation of the pros and cons of the two approaches when applied to realistic industry use cases.

Collaborative Generative Hashing for Marketing and Fast Cold-Start Recommendation

Cold-start has being a critical issue in recommender systems with the explosion of data in e-commerce. Most existing studies proposed to alleviate the cold-start problem are also known as hybrid recommender systems that learn representations of users and items by combining user-item interactive and user/item content information. However, previous hybrid methods regularly suffered poor efficiency bottlenecking in online recommendations with large-scale items, because they were designed to project users and items into continuous latent space where the online recommendation is expensive. To this end, we propose a collaborative generated hashing (CGH) framework to improve the efficiency by denoting users and items as binary codes, then fast hashing search techniques can be used to speed up the online recommendation. In addition, the proposed CGH can generate potential users or items for marketing application where the generative network is designed with the principle of minimum description length, which is used to learn compact and informative binary codes. Extensive experiments on two public datasets show the advantages for recommendations in various settings over competing baselines and analyze its feasibility in marketing application.

Research on Trust Evaluation Model Based on Statistical Data in E-Commerce

<p>Both the level of science and technology and people’s economic level have been significantly improved in the present social background. With the application and popularization of computers and networks in people’s homes, online shopping, which is of high flexibility and convenience, has emerged and gradually become an important choice to shop. In e-commerce activities, the trust between the two parties often directly affects the results of online transactions, which is even a very critical influencing factor. Therefore, an effective trust evaluation model will make a positive impact on the healthy and sustainable development of e-commerce activities. Based on the trust evaluation model of reputation that has been applied in practice, this article will integrate D-S evidence fusion algorithm into the new evaluation model and try to analyze its unique value in application.</p>

Survey on Recommendation of E-commerce products to existing and new user by Analyzing Social data along with e-commerce data

Survey on Recommendation of E-commerce products to existing and new user by Analyzing Social data along with e-commerce data

Estimating city-level poverty rate based on e-commerce data with machine learning

There are many big data sources in Indonesia, for example, data from social media, financial transactions, transportation, call detail records, and e-commerce. These types of data have been considered as potential resources to complement periodic surveys and censuses to monitor development indicators such as poverty levels. Data from e-commerce in particular could potentially represent the real expenditure of households, better complying with the formal calculation of the poverty line than other datasets. The contribution of this research is to propose a framework for poverty rate estimation based on e-commerce data using machine learning algorithms. The influence of items and aspects in e-commerce data was investigated in conjunction with poverty rate estimation. The experimental result showed that e-commerce data could potentially be used as a proxy for calculating city-level poverty rates. It was also found that cars and motorbikes are the two most significant items for poverty prediction in Indonesia.

The Role of Perceived Risk and Trust Propensity in The Relationship Between Negative Perceptions of Applying Big Data Analytics and Consumers’ Responses

With the phenomenal growth of Big Data in e-commerce, applying big data analytics brings negative perception for customers, in one way or another. The research on negative perception of applying big data analytics and the role of perceived risk and trust propensity to consumers’ responses under applying Big Data analytics is lacking. Therefore, the aims of this study are to analyze the role of perceived risk and trust propensity in the relationship between negative perceptions of applying big data analytics and consumers’ responses. A sample of 349 respondents was used in data analysis. The study found out that perceived risk don’t act mediate the relationship between negative perception of applying BDA and consumers’ responses. Besides, customers’ trust propensity was found to moderate the relation of negative perception of applying BDA to customers’ responses and perceived risk to customers’ responses. High trust propensity participants reported stronger responses than those with low trust propensity. It due to customers’ trust on new applications of BDA, hence, it is easy to influence on customers as their negative response when negative perception and perceived risk are rising. The findings of this research will have implications for e-vendors to understand the important role of perceived risk and trust propensity on customers’ responses under Big Data analytics era.

Quantitative Evaluation of Sentiment Analysis on E-Commerce Data

With the emergence of technology and Big Data in the modern world, it is important to govern such vast amounts of data. Amazon and other e-commerce websites play a vital role in the delivery of goods services to the user. Nonetheless, such services are frequently accompanied by consumer reviews and ratings of the things they sell. Such reviews and ratings provided in the form of textual feedback serve to improve service delivery and product quality in the event that a consumer is dissatisfied. Thus, the process of evaluations and ratings is regarded as a significant component of the customer satisfaction process. Thus, it is vital to analyse them so that useful insights can be gleaned from them and the correct judgements can be made regarding the enhancement of the product. With emphasis on this concept, the authors of the proposed paper intend to develop a model capable of performing sentiment analysis on customer-provided product reviews and analysing products with positive and negative comments. On the basis of these reviews, three machine learning algorithms, primarily Decision Trees, XGBoost, and AdaBoost, are utilised to undertake sentiment analysis on products. It was noticed that XGBoost produced a greater detection accuracy of 84.45%.

Study on convolutional neural network and its application in data mining and sales forecasting for E-commerce

In recent years, the rapid development of e-commerce has brought great convenience to people. Compared with traditional business environment, e-commerce is more dynamic and complex, which brings many challenges. Data mining technology can help people better deal with these challenges. Traditional data mining technology cannot effectively use the massive data in the electricity supplier, it relies on the time-consuming and labour-consuming characteristic engineering, and the obtained model is not scalable. Convolutional neural network can effectively use a large amount of data, and can automatically extract effective features from the original data, with higher availability. In this paper, convolutional neural network is used to mine e-commerce data to achieve the prediction of commodity sales. First, this article combines the inherent nature of the relevant merchandise information with the original cargo log data that can be converted into a specific “data frame” format. Raw log data includes items sold over a long period of time, price, quantity view, browse, search, search, times collected, number of items added to cart, and many other metrics. Then, convolutional neural network is applied to extract effective features on the data frame. Finally, the final layer of the convolutional neural network uses these features to predict sales of goods. This method can automatically extract effective features from the original structured time series data by convolutional neural network, and further use these features to achieve sales forecast. The validity of the proposed algorithm is verified on the real e-commerce data set.

Customer gender prediction system on hierarchical E-commerce data

BackgroundE-commerce services provide online shopping sites and mobile applications for small and medium sellers. To provide more efficient buying and selling experiences, a machine learning system can be applied to predict the optimal organization and display of products that maximize the chance of bringing useful information to user that facilitate the online purchases. Therefore, it is important to understand the relevant products for a gender to facilitate the online purchases. In this work, we present a statistical machine learning (ML)-based gender prediction system to predict the gender “male” or “female” from transactional E-commerce data. We introduce different sets of learning algorithms including unique IDs decomposition, context window-based history generation, and extract identical hierarchy from training set to address the gender prediction classification system from online transnational data.ResultsThe experiment result shows that different feature augmentation approaches as well as different term or feature weighting approaches can significantly enhance the performance of statistical machine learning-based gender prediction system.ConclusionsThis work presents a ML-based implementational approach to address E-commerce-based gender prediction system. Different session augmentation approaches with support vector machines (SVMs) classifier can significantly improve the performance of gender prediction system.

Predicting online shopping behaviour from clickstream data using deep learning

Clickstream data is an important source to enhance user experience and pursue business objectives in e-commerce. The paper uses clickstream data to predict online shopping behavior and target marketing interventions in real-time. Such AI-driven targeting has proven to save huge amounts of marketing costs and raise shop revenue. Previous user behavior prediction models rely on supervised machine learning (SML). Conceptually, SML is less suitable because it cannot account for the sequential structure of clickstream data. The paper proposes a methodology capable of unlocking the full potential of clickstream data using the framework of recurrent neural networks (RNNs). An empirical evaluation based on real-world e-commerce data systematically assesses multiple RNN classifiers and compares them to SML benchmarks. To this end, the paper proposes an approach to measure the revenue impact of a targeting model. Estimates of revenue impact together with results of standard classifier performance metrics evidence the viability of RNN-based clickstream modeling and guide employing deep recurrent learners for campaign targeting. Given that the empirical analysis shows RNN-based and conventional classifiers to capture different patterns in clickstream data, a specific recommendation is to combine sequence and conventional classifiers in an ensemble. The paper shows such an ensemble to consistently outperform the alternative models considered in the study.