Purchase Prediction Research Articles

Product Recommendation Based on Eye Tracking Data Using Fixation Duration

E-commerce can be used to increase companies or sellers’ profits. For consumers, e-commerce can help them shop faster. The weakness of e-commerce is that there is too much product information presented in the catalog which in turn makes consumers confused. The solution is by providing product recommendations. As the development of sensor technology, eye tracker can capture user attention when shopping. The user attention was used as data of consumer interest in the product in the form of fixation duration following the Bojko taxonomy. The fixation duration data was processed into product purchase prediction data to know consumers’ desire to buy the products by using Chandon method. Both data could be used as variables to make product recommendations based on eye tracking data. The implementation of the product recommendations based on eye tracking data was an eye tracking experiment at selvahouse.com which sells hijab and women modest wear. The result was a list of products that have similarities to other products. The product recommendation method used was item-to-item collaborative filtering. The novelty of this research is the use of eye tracking data, namely the fixation duration and product purchase prediction data as variables for product recommendations. Product recommendation that produced by eye tracking data can be solution of product recommendation’s problems, namely sparsity and cold start.

Factorizing Historical User Actions for Next-Day Purchase Prediction

It is common practice for many large e-commerce operators to analyze daily logged transaction data to predict customer purchase behavior, which may potentially lead to more effective recommendations and increased sales. Traditional recommendation techniques based on collaborative filtering, although having gained success in video and music recommendation, are not sufficient to fully leverage the diverse information contained in the implicit user behavior on e-commerce platforms. In this article, we analyze user action records in the Alibaba Mobile Recommendation dataset from the Alibaba Tianchi Data Lab, as well as the Retailrocket recommender system dataset from the Retail Rocket website. To estimate the probability that a user will purchase a certain item tomorrow, we propose a new model called Time-decayed Multifaceted Factorizing Personalized Markov Chains (Time-decayed Multifaceted-FPMC), taking into account multiple types of user historical actions not only limited to past purchases but also including various behaviors such as clicks, collects and add-to-carts. Our model also considers the time-decay effect of the influence of past actions. To learn the parameters in the proposed model, we further propose a unified framework named Bayesian Sparse Factorization Machines. It generalizes the theory of traditional Factorization Machines to a more flexible learning structure and trains the Time-decayed Multifaceted-FPMC with the Markov Chain Monte Carlo method. Extensive evaluations based on multiple real-world datasets demonstrate that our proposed approaches significantly outperform various existing purchase recommendation algorithms.

Sales and Stock Purchase Prediction System Using Trend Moment Method and FIS Tsukamoto

The purpose of this research is to build decisions support model to predict sales and stock purchase using Trend Moment method and Tsukamoto Fuzzy Inference System. Trend moment is a simple statistical-based forecasting method widely used to forecast sales in a company using historical data. Tsukamoto is a fuzzy inference system that uses monotonic reasoning to determine output. The object of this research is sales and purchase data for Ice Cream X Depo Jambi products from August 2019 to April 2020. The study aims to build a decision support model web-based to predict sales and purchases of ice cream X stock at Jambi depots. Fuzzy Tsukamoto in this study will be used to predict product stock purchases after predicting future sales using trend moments. The system input is in product data form, data of ice cream sales history, and data of ice cream stock purchase. Sales history data will be use to calculate slope and constanta that will predict future sales trends. Stocks purchase history data along with sales trend prediction value will be use to calculate the membership degree of fuzzy variables, perform the aggregation process on fuzzy rules, and then carry out the defuzzification process to produce output prediction values for future ice cream stock purchases. from the prediction model implemented in the decision support system, sales prediction data has an accuracy of 71% while stock purchase predictions have an accuracy of 85%.

On the platform but will they buy? Predicting customers' purchase behavior using deep learning

A thorough understanding of online customer's purchase behavior will directly boost e-commerce business performance. Existing studies have overtly focused on purchase intention and used sales rank as a natural proxy, which however has limited business application. Additionally, intention to purchase does not necessarily convert to actual retail purchases. We aim to further our understanding of online customer's purchase behavior for an e-commerce platform by predicting the same using deep learning techniques, on a large multidimensional data sample of more than 50,000 unique web sessions. This study used two distinct sets of variables, i.e., platform engagement and customer characteristics, as key predictors of online purchases by retail customers. We further compared the predictive capability of our deep learning method with other widely used machine learning techniques for prediction, including Decision Tree, Random Forest, Support Vector Machines, and Artificial Neural Networks. We found that the deep learning technique outperformed the machine learning techniques when applied to the same dataset. These analyses will help platform designers plan for more platform engagements while simultaneously expanding the academic understanding of purchase prediction for online e-commerce platforms.

Predicting agricultural and livestock products purchases using the Internet search index and data mining techniques

PurposeThis study identifies whether the Internet search index can be used as effective enough data to identify agricultural and livestock product demand and compare the accuracy of the prediction of major agricultural and livestock products purchases between these prediction models using artificial neural network, linear regression and a decision tree.Design/methodology/approachArtificial neural network, linear regression and decision tree algorithms were used in this study to compare the accuracy of the prediction of major agricultural and livestock products purchases. The analysis data were studied using 10-fold cross validation.FindingsFirst, the importance of the Internet search index among the 20 explanatory variables was found to be high for most items, so the Internet search index can be used as a variable to explain agricultural and livestock products purchases. Second, as a result of comparing the accuracy of the prediction of six agricultural and livestock purchases using three models, beef was the most predictable, followed by radishes, chicken, Chinese cabbage, garlic and dried peppers, and by model, a decision tree shows the highest accuracy of prediction, followed by linear regression and an artificial neural network.Originality/valueThis study is meaningful in that it analyzes the purchase of agricultural and livestock products using data from actual consumers' purchases of agricultural and livestock products. In addition, the use of data mining techniques and Internet search index in the analysis of agricultural and livestock purchases contributes to improving the accuracy and efficiency of agricultural and livestock purchase predictions.

Robust Spatio-Temporal Purchase Prediction via Deep Meta Learning

Purchase prediction is an essential task in both online and offline retail industry, especially during major shopping festivals, when strong promotion boosts consumption dramatically. It is important for merchants to forecast such surge of sales and have better preparation. This is a challenging problem, as the purchase patterns during shopping festivals are significantly different from usual cases and also rare in historical data. Most existing methods fail at this problem due to the extremely scarce data samples as well as the inability to capture the complex macroscopic spatio-temporal dependencies in a city. To address this problem, we propose the Spatio-Temporal Meta-learning Prediction (STMP) model for purchase prediction during shopping festivals. STMP is a meta-learning based spatio-temporal multi-task deep generative model. It adopts a meta-learning framework with few-shot learning capability to capture both spatial and temporal data representations. A generative component then uses the extracted spatio-temporal representation and input data to infer the prediction results. Extensive experiments demonstrate the meta-learning generalization ability of STMP. STMP outperforms baselines in all cases, which shows the effectiveness of our model.

Customer purchase prediction from the perspective of imbalanced data: A machine learning framework based on factorization machine

Customer purchase prediction aims to predict customers' future purchases, and the prediction results are of great importance for conducting future commercial activities. To obtain accurate predictions of customer purchases, this paper develops a machine learning framework based on historical behavioural data. First, considering the sparsity of behavioural data, this paper proposes a feature combination method based on the improved factorization machine algorithm. Second, due to the imbalance of customer purchase data, this paper proposes an imbalanced prediction method based on the maximized marginal category and cost-sensitive ensemble learning. Finally, a real-word travel service purchase dataset is adopted to test the feasibility of the proposed prediction framework. The experimental results and comparative analysis verify the validity of the proposed model.

Towards purchase prediction: A transaction-based setting and a graph-based method leveraging price information

Targeting at boosting business revenue, purchase prediction based on user behavior is crucial to e-commerce. However, it is not a well-explored topic due to a lack of relevant datasets. Specifically, no public dataset provides both price and discount information varying on time, which play an essential role in the user’s decision making. Besides, existing learn-to-rank methods cannot explicitly predict the purchase possibility for a specific user-item pair. In this paper, we propose a two-step graph-based model, where the graph model is applied in the first step to learn representations of both users and items over click-through data, and the second step is a classifier incorporating the price information of each transaction record. To evaluate the model performance, we propose a transaction-based framework focusing on the purchased items and their context clicks, which contain items that a user is interested in but fails to choose after comparison. Our experiments show that exploiting the price and discount information can significantly enhance prediction accuracy.

Towards early purchase intention prediction in online session based retailing systems

Purchase prediction has an important role for decision-makers in e-commerce to improve consumer experience, provide personalised recommendations and increase revenue. Many works investigated purchase prediction for session logs by analysing users’ behaviour to predict purchase intention after a session has ended. In most cases, e-shoppers prefer to be anonymous while browsing the websites and after a session has ended, identifying users and offering discounts can be challenging. Therefore, after a session ends, predicting purchase intention may not be useful for the e-commerce strategists. In this work, we propose and develop an early purchase prediction framework using advanced machine learning models to investigate how early purchase intention in an ongoing session can be predicted. Since users could be anonymous, this could help to give real-time offers and discounts before the session ends. We use dynamically created session features after each interaction in a session, and propose a utility scoring method to evaluate how early machine learning models can predict the probability of purchase intention. The proposed framework is validated with a real-world dataset. Computational experiments show machine learning models can identify purchase intention early with good performance in terms of Area Under Curve (AUC) score which shows success rate of machine learning models on early purchase prediction.

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.

Product Quantization using Regression

Approximate Nearest Neighbor (ANN) has developed an immense demand for many tasks. This ANN methodology was being used for product quantization. These product quantization methods were being used for e-commerce sites. However, this quantization maybe sometimes misleading due to a lack of accuracy in technique. So, we managed to increase the accuracy of quantization by adding Logistic Regression in the process. This helps to increase the accuracy of the method by having a probability value. This helps to make correlated items much more accurate when compared to pure quantization. This method is helpful for e-commerce sites for efficiency in the prediction of purchase by the customer.

Click-aware purchase prediction with push at the top

Eliciting user preferences from purchase records for performing purchase prediction is challenging because negative feedback is not explicitly observed, and because treating all non-purchased items equally as negative feedback is unrealistic. Therefore, in this study, we present a framework that leverages the past click records of users to compensate for the missing user–item interactions of purchase records, i.e., non-purchased items. We begin by formulating various model assumptions, each one assuming a different order of user preferences among purchased, clicked-but- not-purchased, and non-clicked items, to study the usefulness of leveraging click records. We implement the model assumptions using the Bayesian personalized ranking model, which maximizes the area under the curve for bipartite ranking. However, we argue that using click records for bipartite ranking needs a meticulously designed model because of the relative unreliableness of click records compared with that of purchase records. Therefore, we ultimately propose a novel learning-to-rank method, called P3Stop, for performing purchase prediction. The proposed model is customized to be robust to relatively unreliable click records by particularly focusing on the accuracy of top-ranked items. Experimental results on two real-world e-commerce datasets demonstrate that P3STop considerably outperforms the state-of-the-art implicit-feedback-based recommendation methods, especially for top-ranked items.

Recommender System for Items in Dota 2

Dota 2 is one of several multiplayer online battle arena games that have recently become extremely popular. A central feature of Dota 2 is that players select and purchase items that are used in the game and the selections strongly affect the outcome of the game. Item recommendations and purchase predictions in Dota 2 turn out to be interesting problems due to the many factors affecting the choice of items and due to the frequency by which items are purchased throughout the game. We present three recommender systems for recommending which items a player might buy throughout a match, based on commonly used purchasing strategies. These systems are a rule-based system, a logistic regression based system, and a logistic regression based system enhanced with clustering. Knowing only the player's hero and inventory items, the first two systems are able to predict the purchases by highly skilled players with accuracies ranging from 66.5% to 87.1%, depending on the hero where accuracy is defined as recommended items that were actually purchased in the next 5 min of the game. For the third system, the players are clustered by purchasing strategy which in turn group players by role and play style. Adding the clustering feature to the second version of the recommender system only improve the results slightly. The reason for this could be that the items in the player's inventory are already a strong indicator of a player's role and play style both of which were used to develop the first two recommender systems. An important question for future work is how the recommendations could be useful in practice for human or AI players.

Exploratory Data Analysis using Python

Data need to be analyzed so as to produce good result. Using the result decision can be taken. For example recommendation system, ranking of the page, demand fore casting, prediction of purchase of the product. There are some leading companies where the review of the customer plays a great role to analyze the factor which influences the review rating. We have used exploratory data analysis (EDA) where data interpretations can be done in row and column format. We have used python for data analysis. it is object oriented ,interpreted and interactive programming language. it is open source with rich sets of libraries like pandas, MATplotlib, seaborn etc. We have used different types of charts and various types of parameter to analyze Amazon review data sets which contains the reviews of electronic data items. We have used python programming for the data analysis.

Online purchase decisions for tourism e-commerce

Abstract In this study, we consider the purchase prediction problem in the context of e-tourism, an emerging and prevailing application in e-commerce. Although a wide array of studies have been taken on purchase prediction, little analysis has been done on the purchasing behaviors towards tourism products. Also, the design of the corresponding purchase prediction model deserves researchers’ full attention. We begin by introducing a real-life e-tourism dataset and constructing a suite of variables based on the detailed current and historical clickstream information. To validate the effectiveness of variables, we then perform a quantitative analysis to address quite a few interesting characteristics of purchase patterns. To predict whether or not a purchase is made for a current visiting session, we present a novel model called co-EM Logistic Regression (co-EM-LR) which combines the semi-supervised learning and the multi-view learning into its procedure. The co-EM-LR model has at least two outstanding merits: (1) it inherits the ease interpretation of the logistic modeling; and (2) it fully exploits both unlabeled data and the compatibility of multiple views to improve the prediction accuracy. Comprehensive experiments demonstrate the proposed co-EM-LR model yields significant prediction performance advantages over five competitive methods. Furthermore, two complementary views can mutually improve the performance with each other and finally offer fast convergence.

An encoder–decoder switch network for purchase prediction

Users in e-commerce tend to click on items of their interest. Eventually, the more frequently an item is clicked by a user, the more likely the item will be purchased by the user after all. However, what if a user clicked on every item only once before purchases? This is a frequently observed user behavior in reality, but predicting which of the clicked items will be purchased is a challenging task. This paper addresses a practical yet widely overlooked task of predicting purchase items within a non-duplicate click session, i.e., a session in which every item is clicked only once. We propose an encoder–decoder neural architecture to simultaneously model users’ click and purchase behaviors. The encoder captures a user’s intent contained in the user’s click session, and the decoder, which is equipped with pointer network via a switch gate, extracts relevant clicked items for future purchase candidates. To the best of our knowledge, our work is the first to address the task of purchase prediction given non-duplicate click sessions. Experiments demonstrate that our proposed method outperforms the state-of-the-art purchase prediction methods by up to 18% in terms of recall.

Will this session end with a purchase? Inferring current purchase intent of anonymous visitors

Abstract Understanding the online behavior and intent of online visitors is the subject of a long line of research. mechanisms to understand the purchase intent of visitors, to increase the number of visits that end with a purchase. anonymous visitors garner little attention, having no shopping history or known interests. Compared to profiled returning customers whose history is known, anonymous visitors garner less attention. The lack of a known shopping history or interests makes it hard to learn from their behavior, or infer their shopping intent. Here, we suggest the use of products’ popularity trends and visit’s temporal information to infer the purchase intention of anonymous visitors. We model these dynamics and utilize our model to infer purchase intent of visitors of two large real e-commerce retailer sites. Our model identifies online signals for purchase intent that can be used for online purchase prediction.

基于数据优化的保险客户承保预测

基于数据优化的保险客户承保预测

Brand purchase prediction based on time‐evolving user behaviors in e‐commerce

SummaryPurchase prediction is a key function in the e‐commerce recommendation system. Existing works usually focus on item‐level purchase prediction, which faces two issues of high cost and low accuracy. In this paper, we study brand purchase prediction by exploring behaviors, which may lead to brand purchases. We make three progresses. (1) We analyze a real world e‐commerce data from multiple angles. Focusing on users' brand purchases, we find behaviors' evolution with time and behaviors' interaction. (2) For different behaviors, we extract different time‐evolving features that can serve as indicators of users' brand purchase. (3) We use a logistic regression‐based model by adjusting the parameters of time‐evolving feature and others in two different scenarios (the promotion purchase prediction and the daily purchase prediction) to construct two experiments. The experiment results show that the model using three types features performs the best in both scenarios, and the time‐evolving feature plays the most important role among them. (4) We distinguish the feature importance in different scenarios. Based on the importance, we find that users' purchases in the promotion scenario are likely to be impulsive, while purchases in the daily scenario are more likely to be influenced by users' activities.

User behaviour modeling, recommendations, and purchase prediction during shopping festivals

This work investigates user online browsing and purchasing behaviors, and predicts purchasing actions during a large shopping festival in China. To improve online shopping experience for consumers, increase sales for merchants and achieve effective warehousing and delivery, we first analyse diverse online shopping behaviours based on the 31 million logs generated accompanied with online shopping during a rushed sale event on 11st November, 2016. Based on the obtained user behaviours and massive data, we apply collaborative filtering based method to recommend items for different consumers, and predict whether purchase will happen. We conduct 5-fold cross validation to evaluate the collaborative filtering based recommendation method, and further identify the critical shopping behaviors that determine the precursors of purchases. As online shopping becomes a global phenomenon, findings in this study have implications on both shopping experience and sales enhancement.