Web Browsing Behavior Research Articles

Privately vertically mining of sequential patterns based on differential privacy with high efficiency and utility

Sequential pattern mining is one of the fundamental tools for many important data analysis tasks, such as web browsing behavior analysis. Based on frequent patterns, decision-makers can obtain both economic gains and social values. Sequential data, on the other hand, frequently contain sensitive information, and directly analyzing these data will raise user concerns from a privacy perspective. Differential privacy (DP), as the most popular privacy model, has been employed to address this privacy concern. Most existing DP-Solutions are designed to combine horizontal sequence pattern mining algorithms with differential privacy. Due to the inefficiency of horizontal algorithms, their DP-Solutions cannot ensure high efficiency and accuracy while offering a high privacy guarantee. Therefore, we proposed privVertical, a new private sequence pattern mining scheme combining the vertical mining algorithm with differential privacy to achieve the above objective. Unlike DP-solutions based on horizontal algorithms, privVertical can promote efficiency by avoiding performing costly database scans or costly projection database constructions. Moreover, to promote accuracy, a differentially private hash MapList (called privHashMap) is designed to record frequent concurrency items and their noisy support based on the Sparse Vector Technique. PrivHashMap is used to pre-pruning excessive infrequent candidate sequences in private mining, and Sparse Vector Technique is used to promote the accuracy of PrivHashMap. After pruning these invalid candidate sequences, less noise is required to achieve the same level of privacy, increasing the accuracy of private mining. Theoretical privacy analysis proves privVertical satisfies varepsilon-differential privacy. Experiments show that privVertical achieves higher accuracy and efficiency while achieving the same privacy level.

User Profiling Behaviour and Activity of Persistent Surveillance of User Activity in Complex Networks

The objective of this paper is to study the behaviour of multiple user activity on network traffic. The traditional approach of using the source IP address as a tracking identifier does not work when used to identify mobile users. Insider misuse has become a significant issue for organizations. Traditional information security has focused on threats from outside rather than employees. A wide range of concepts has been undertaken to develop approaches to detecting the insider. Recent efforts to address this problem by exclusively not relying on web browsing behaviour, collecting data from routers, and predicting human psychology (when a router receives a packet, the router checks its routing table to determine if the table lists the destination address in the header. If the table does not contain the destination address, the router forwards the packet to another router that is listed in its routing table). This paper builds the concept of tracking behaviour-based identification of multiple user activity on a router. Here we find out the traffic of HTTP or HTTPS Protocol and the Results are promising. find out as the result we are capable of tracking human psychology.

The role of web browsing in credit risk prediction

Online mail order and online retail purchases have increased rapidly in recent years worldwide, with Covid-19 forcing almost all non-grocery shopping to move online. These practices have facilitated the availability of new data sources, such as web behavioural variables providing scope for innovation in credit risk analysis and decision practices. This paper examines new web browsing variables and incorporates them into survival analysis as predictors of probability of default (PD). Using a large sample of purchase and repayment credit accounts from a major digital retailer and financial services provider, we show that these new variables enhance the predictive accuracy of probability of default (PD) models at account level. This also holds in the absence of credit bureau data, therefore, the new information can help people who may not have a credit history (thin file) who cannot be assessed using traditional variables. Moreover, we leverage on the dynamic nature of these new web variables and explore their predictive value in short and long- term horizons. By adding macroeconomic variables, the possibility for stress-testing is provided. Our empirical findings provide insights into web browsing behaviour, highlight how the inclusion of non-standard variables can improve credit risk scoring models and lending decisions and may provide a solution to the thin files problem. Our results also suggest a direct value added to the online retail credit industry as firms should leverage the increasing trend of consumers embracing the digital environment.

A Fusion Model Based on Dynamic Web Browsing Behavior Analysis for IoT Insider Threat Detection.

With the wide application of Internet of things (IoT) devices in enterprises, the traditional boundary defense mechanisms are difficult to satisfy the demands of the insider threats detection. IoT insider threat detection can be more challenging, since internal employees are born with the ability to escape the deployed information security mechanism, such as firewalls and endpoint protection. In order to detect internal attacks more accurately, we can analyze users’ web browsing behaviors to identify abnormal users. The existing web browsing behavior anomaly detection methods ignore the dynamic change of the web browsing behavior of the target user and the behavior consistency of the target user in its peer group, which results in a complex modeling process, low system efficiency and low detection accuracy. Therefore, the paper respectively proposes the individual user behavior model and the peer-group behavior model to characterize the abnormal dynamic change of user browsing behavior and compare the mutual behavioral inconsistency among one peer-group. Furthermore, the fusion model is presented for insider threat detection which simultaneously considers individual behavioral abnormal dynamic changes and mutual behavioral dynamic inconsistency from peers. The experimental results show that the proposed fusion model can accurately detect insider threat based on the abnormal user web browsing behaviors in the enterprise networks.

Nowcasting Entrepreneurship: Urban Third Place versus the Creative Class

Researchers have long debated whether entrepreneurship policy should focus on place or people. In this paper, we extend the place-based versus people-based theories using contemporaneous and geographically granular web-user online activity data to predict a region’s proclivity for entrepreneurship. We compare two theoretical hypotheses: the urban third places—informal gathering locations—that facilitate social interaction and entrepreneurship, in contrast to the creative class which fosters entrepreneurial energy and opportunity in a region. Specifically, we assess whether business formation has a stronger statistical relationship with the browsing behavior of individuals visiting websites associated with third place locations—e.g., restaurants or bars—or the concentration of web browsing behavior associated with “the creative class”. Using U.S. county-level data, we find that both urban third places and the creative class can predict about 70% of the variations in regional business formation, with the creative class having a slight competitive edge.

Reinforcement Learning Page Prediction for Hierarchically Ordered Municipal Websites

Public websites offer information on a variety of topics and services and are accessed by users with varying skills to browse the kind of electronic document repositories. However, the complex website structure and diversity of web browsing behavior create a challenging task for click prediction. This paper presents the results of a novel reinforcement learning approach to model user browsing patterns in a hierarchically ordered municipal website. We study how accurate predictor the browsing history is, when the target pages are not immediate next pages pointed by hyperlinks, but appear a number of levels down the hierarchy. We compare traditional type of baseline classifiers’ performance against our reinforcement learning-based training algorithm.

Understanding perceived barriers to treatment from web browsing behavior

BackgroundThe expanding amount of information available from our use of technologies has led researchers to explore how this information can aid in the detection of mental health issues. We expand on past work in this area by exploring how browsing histories might be able to predict perceived barriers to psychological treatment. MethodsWe obtained 10 days of browsing history data for 255 respondents as well as assessments of Perceived Barriers to Psychological Treatments and depression, the Patient Health Questionnaire. ResultsWe found that browsing histories enabled high performance classification of people with high levels of perceived barriers to psychological treatments (AUC average of 0.86). LimitationsOur high classification accuracy does not help understand why different features within the browsing histories are useful to classify people according to browsing history. We also look at people who decided to contribute their browsing history but the use of this data more generally presents additional ethical questions. ConclusionsBrowsing histories might be useful to classify people's barriers to seeking psychological treatment. It is clinically relevant to find those who perceive barriers to seeking treatment to better design ways to address those concerns and help them find treatment.

Siamese Neural Networks for User Identity Linkage Through Web Browsing.

Linking online identities of users among countless heterogeneous network services on the Internet can provide an explicit digital representation of users, which can benefit both research and industry. In recent years, user identity linkage (UIL) through the Internet has become an emerging task with great potential and many challenges. Existing works mainly focus on online social networks that consider inconsistent profiles, content, and networks as features or use sparse location-based data sets to link the online behaviors of a real person. To extend the UIL problem to a general scenario, we try to link the web-browsing behaviors of users, which can help to distinguish specific users from others, such as children or malicious users. More specifically, we propose a Siamese neural network (NN) architecture-based UIL (SAUIL) model that learns and compares the highest-level feature representation of input web-browsing behaviors with deep NNs. Although the number of matching and nonmatching pairs for the UIL problem is highly imbalanced, previous studies have not considered imbalanced UIL data sets. Therefore, we further address the imbalanced learning issue by proposing cost-sensitive SAUIL (C-SAUIL) model, which assumes higher costs for misclassifying the minority class. In the experiments, the proposed model is robust and exhibits a good performance on very large, real-world data sets collected from different regions with distinct characteristics.

Behavior Analysis for Electronic Commerce Trading Systems: A Survey

With the rapid development of the Internet and the continuous progress of computer science and technology, Internet-based online trading has developed rapidly and improved the convenience of everyday life. Increasingly, people conduct business activities through online transactions and payment methods, and the development prospects for online transactions are quite broad. In addition, e-commerce trading systems have attracted considerable attention. Many studies have been conducted to analyze the behaviors of these systems. This paper focuses on analyzing both system behavior and user behavior in e-commerce trading systems. To reflect the characteristics of online trading systems, system behavior reflects the strength of the system, while user behavior reflects various user actions. Consistency, which is a basis for determining whether two objects are similar, provides a reference for designers and modelers. This paper also focuses on analyzing the consistency of system behavior, especially regarding the measurement and diagnosis of e-commerce trading systems. Then, we study the existing user behavior analysis methods for e-commerce trading systems. We divide user behaviors into four groups: web browsing behavior, keystroke behavior, network transaction behavior, and mobile terminal behavior. The performance and correlation analysis techniques of these methods are discussed and compared in detail. Finally, we provide ideas for future research involving behavior analysis for online trading systems.

Browsing Behaviour Analysis using Data Mining

Now-a-days most of our time is spent online using some form of digital technology such as search engines, news portals, or social media websites. Our online presence makes us engaged most of the time and leads us to become oblivious of our important work, resulting in a form of procrastination that decreases our productivity significantly. Some desktop and mobile applications have recently emerged to counter the problem by introducing various means of self-tracking to reduce the wasting of time and engage in productive activities. However, these systems suffer several shortcomings in terms of being static or providing a limited view of actions using one aspect only. To promote self-awareness that helps bring positive changes in individual’s performance, there is a need to present the data in a more persuasive ways, bringing interaction to it and present the same data in different ways using both temporal and cate-gorical dimensions. We describe a framework that collects and processes the browsing data and creates a user behavior model to extract valuable and interesting temporal and categorical patterns regarding user online behavior and interests. To discover the valuable behavior patterns from the individual’s browsing data, different web usage mining techniques have been used. Finally, we demonstrate interactive visualizations for the analysis and monitoring of web browsing behavior patterns with the goal of providing the individual with detailed understanding of his/her behavior. We also present a small-scale study including university students, which proves the importance of our work.

Predicting demographics, moral foundations, and human values from digital behaviours

Personal electronic devices including smartphones give access to behavioural signals that can be used to learn about the characteristics and preferences of individuals. In this study, we explore the connection between demographic and psychological attributes and the digital behavioural records, for a cohort of 7633 people, closely representative of the US population with respect to gender, age, geographical distribution, education, and income. Along with the demographic data, we collected self-reported assessments on validated psychometric questionnaires for moral traits and basic human values, and combined this information with passively collected multi-modal digital data from web browsing behaviour and smartphone usage. A machine learning framework was then designed to infer both the demographic and psychological attributes from the behavioural data. In a cross-validated setting, our models predicted demographic attributes with good accuracy as measured by the weighted AUROC score (Area Under the Receiver Operating Characteristic), but were less performant for the moral traits and human values. These results call for further investigation, since they are still far from unveiling individuals' psychological fabric. This connection, along with the most predictive features that we provide for each attribute, might prove useful for designing personalised services, communication strategies, and interventions, and can be used to sketch a portrait of people with similar worldview.

Bayes, not Naïve: Security Bounds on Website Fingerprinting Defenses

Abstract Website Fingerprinting (WF) attacks raise major concerns about users’ privacy. They employ Machine Learning (ML) techniques to allow a local passive adversary to uncover the Web browsing behavior of a user, even if she browses through an encrypted tunnel (e.g. Tor, VPN). Numerous defenses have been proposed in the past; however, it is typically difficult to have formal guarantees on their security, which is most often evaluated empirically against state-of-the-art attacks. In this paper, we present a practical method to derive security bounds for any WF defense, where the bounds depend on a chosen feature set. This result derives from reducing WF attacks to an ML classification task, where we can determine the smallest achievable error (the Bayes error). Such error can be estimated in practice, and is a lower bound for a WF adversary, for any classification algorithm he may use. Our work has two main consequences: i) it allows determining the security of WF defenses, in a black-box manner, with respect to the state-of-the-art feature set and ii) it favors shifting the focus of future WF research to identifying optimal feature sets. The generality of this approach further suggests that the method could be used to define security bounds for other ML-based attacks.

The Digital Revolution, 3D Printing, and Innovation as Data

The digital revolution has created a data‐rich environment. Increasingly, firms are seeking to acquire and analyze a variety of consumer data such as online shopping, social media, and web browsing behavior to enhance their innovation activities. This approach is termed “Innovation from Data.” At the same time, a growing number of consumers are gaining the ability to transform digital data into innovative physical products through the use of new tools such as audio and video editing software and 3D printing technology. This approach is termed “Innovation as Data.” These two approaches differ substantially in terms of their views on the role of consumers, the relevance of firms, and the nature of products. This article introduces the concept of Innovation as Data and discusses its relation to the digital revolution in general and 3D printing in particular. In addition, it offers a set of recommendations for managers seeking to leverage this new innovation approach and ideas for scholars interested in researching this intriguing domain.

Remote control: evaluating the potential of virtual desktops as a data collection tool in studies exploring how people use the Internet

This study explores how virtual desktops hosted within a cloud computing environment can be used as a data capture tool to undertake research exploring web-browsing and web-searching behaviour. It details the protocol used to observe the task-orientated web interactions of 105 university and school aged participants within the confines of a research experiment and considers the comparability between data collected remotely via virtual desktops and data collected in a controlled environment. Data collected remotely via virtual desktops is found to be comparable to data collected in the field across a number of key statistics, with remote data collection making recruitment and participation more convenient, whilst potentially reducing response bias. The protocol detailed has broad potential for adaption where there is a need to remotely capture participants’ web-use in a cost-effective and non-technical manner. Limitations with the protocol are identified and possibilities for future adaptations are discussed.

The user attribution problem and the challenge of persistent surveillance of user activity in complex networks

In telecommunication networks, the user attribution problem refers to the challenge faced in recognizing communi- cation traffic as belonging to a given user when information needed to identify the user is missing. This problem becomes more difficult to tackle as users move across many mobile networks (complex networks) owned and operated by different providers. The traditional approach of using the source IP address as a tracking identifier does not work when used to identify mobile users. Recent efforts to address this problem by exclusively relying on web browsing behavior to identify users, brought to light the challenges of solutions which try to link up multiple user sessions together when these approaches rely exclusively on the frequency of web sites visited by the user. This study has tackled this problem by utilizing behavior based identification while accounting for time and the sequential order of web visits by a user. Hierarchical Temporal Memories (HTM) were used to classify historical navigational patterns for different users. This approach enables linking multiple user sessions together forgoing the need for a tracking identifier such as the source IP address. Results are promising. HTMs outperform traditional Markov chains based approaches and can provide high levels of identification accuracy.

IJARCCE - Computer and Communication Engineering

Predicting the individual's web-browsing behavior on the internet is more important for advertisement mining. This makes the advertisers or the publishers to successfully interact with users in providing relevant advertisement, many intelligent interfaces requires a method for recognizing, analyzing and predicting user behavior actions. Initially we need to collect a log of datasets with user behavior attributes. Hidden Markov Model is used to derive a pattern for predicting the behavior of the users on the web. Also, it helps in analyzing the performance of user behavior profile and estimating the advertising cost to perform Mapreduce jobs based on user's search query. With these collected information, optimized advertisement is deployed to the user using Bigquery mechanism.

HTTP-sCAN: Detecting HTTP-flooding attack by modeling multi-features of web browsing behavior from noisy web-logs

HTTP-flooding attack disables the victimized web server by sending a large number of HTTP Get requests. Recent research tends to detect HTTP-flooding with the anomaly-based approaches, which detect the HTTP-flooding by modeling the behavior of normal web surfers. However, most of the existing anomaly-based detection approaches usually cannot filter the web-crawling traces from unknown searching bots mixed in normal web browsing logs. These web-crawling traces can bias the base-line profile of anomaly-based schemes in their training phase, and further degrade their detection performance. This paper proposes a novel web-crawling traces-tolerated method to build baseline profile, and designs a new anomaly-based HTTP-flooding detection scheme (abbr. HTTP-sCAN). The simulation results show that HTTP-sCAN is immune to the interferences of unknown web-crawling traces, and can detect all HTTP-flooding attacks.

Decision Fusion for Multimodal Active Authentication

The authors apply a decision fusion architecture on a collection of behavioral biometric sensors using keystroke dynamics, mouse movement, stylometry, and Web browsing behavior. They test this active authentication approach on a dataset collected from 19 individuals in an office environment.

WeBeVis: analyzing user web behavior through visual metaphors

The rapid growth of Internet usage has dramatically changed the way we interact with the outside world. Many people read news, communicate with friends and purchase goods online. These activities are usually done via web browsing, and web browsers record information about these activities. The recorded data can be used to understand web browsing behavior of users and improve their browsing experience. For example, website usability and the personalization of online services could both benefit from knowledge of user browsing behavior. A number of methods including data mining, text processing and visualization have been used to uncover user browsing patterns. However, these methods are mainly used to analyze and gain insights into collective behavior patterns of either a large amount of separate web users or users within an online community over a prolonged period of time. Very few systems are available for analyzing the detailed behavior of a single user within a relatively short and specific period of time. In an attempt to shorten this gap, we have developed a visual analytic system called WeBeVis. This system offers three different ways of visualizing web browsing data based on our proposed visual metaphors. It also provides a common interface for users to interact with the visualizations. In this paper, we describe this system and present a user study of it. We show that by visualizing the web browsing history of a user, we are able to uncover interesting patterns in the way that individuals use the web.

Using Internet Behavior to Deliver Relevant Television Commercials

Consumer footprints left on the Internet help advertisers show consumers relevant Web ads, which increase awareness and click-throughs. This “proof of concept” experiment illustrates how Internet behavior can identify relevant television commercials that increase ad-effectiveness by raising attention and ad exposure. Product involvement and prior brand exposure, however, complicate effective Internet-targeting. Ad relevance matters more for low-involvement products, which have a short pre-purchase search process. For the same reason, using Web browsing behavior to make inferences about current ad relevance is more accurate for low-involvement products. Prior brand exposure reduces information-value, even for relevant commercials, and therefore dampens ad relevance's effect on attention and ad exposure.