User Behavior Analysis Research Articles

Based on intelligent advertising recommendation and abnormal advertising monitoring system in the field of machine learning

With the rapid development of the Internet, the scale of the online advertising market has expanded rapidly, and display advertising has become the most popular means of publicity. Accurate advertising recommendation is the guarantee of Internet platform revenue, and accurate advertising click rate prediction is the premise of accurate recommendation and abnormal advertising detection. Therefore, such monitoring and recommendation can be achieved through machine learning combined with artificial intelligence, and the application of intelligent AD recommendation systems and abnormal AD monitoring in the field of machine learning represents a complex integration of technologies to improve the precision and effectiveness of digital marketing strategies. Intelligent AD recommendation systems utilize advanced machine learning algorithms to analyze user behavior and preferences to deliver tailored AD content. These systems leverage vast amounts of user data, including browsing history, purchase history, and engagement metrics, to predict and present the most relevant ads. This paper analyzes the data mining in machine learning algorithms and the real-time online recommendation algorithm of Gaussian process, and analyzes the abnormal advertising monitoring system for maintaining the integrity and efficiency of advertising campaigns. By using machine learning technology for pattern recognition and anomaly detection, various measures and indicators of advertising campaigns can be monitored in a vigilant manner.

The human factor in phishing: Collecting and analyzing user behavior when reading emails

Phishing emails are constantly increasing their sophistication, and typical countermeasures struggle at addressing them. Attackers target our cognitive vulnerabilities with a varied set of techniques, and each of us, not trained enough or simply in the wrong moment, can be deceived and put an entire organization in trouble. To date, no study has evaluated the behavior of users when confronted with phishing emails characterized by a diverse set of features and attack strategies. We created a system called Spamley from these observations, which has the main aim of collecting and sharing data regarding such user behavior. Spamley is also meant to disseminate awareness among users. To reach these goals, we firstly analyzed the wide scientific literature on phishing. Our analysis shows that the focus of studies on phishing has more and more shifted from technical to human-oriented aspects. Building on this analysis we designed, implemented, and deployed a system comprising a web application to test user awareness about phishing, featuring a survey to identify the most interesting characteristics of users, and fueled by a large and varied set of test emails engineered to solicit the several possible cognitive vulnerabilities we all have. We describe in details the design and implementation choices, the lessons we learned, and the way we filled the gap in the available related work. Finally, we use real data from our first 500 users to show how data collected can be used for several important analyses, including which characteristics of the emails are more relevant for which cognitive vulnerability of specific groups of users. Results obtained can guide the development of novel email clients as well as tailored training programmes. Data collected is available to the scientific community for conducting further studies on the important and still unsolved issue of email phishing.

Measuring digital literacy with eye tracking: an examination of skills and performance based on user gaze

PurposeThis study presents an innovative approach to analyzing user behavior when performing digital tasks by integrating eye-tracking technology. Through the measurement of user scan patterns, gaze and attention during task completion, the authors gain valuable insights into users' approaches and execution of these tasks.Design/methodology/approachIn this research, the authors conducted an observational study that centered on assessing the digital skills of individuals with limited proficiency who enrolled in a computer introductory course. A group of 19 participants were tasked with completing various online assignments both before and after completing the course.FindingsThe study findings indicate a significant improvement in participants' skills, particularly in basic and straightforward applications. However, advancements in more sophisticated utilization, such as mastering efficient search techniques or harnessing the Internet for enhanced situational awareness, demonstrate only marginal enhancement.Originality/valueIn recent decades, extensive research has been conducted on the issue of digital inequality, given its significant societal implications. This paper introduces a novel tool designed to analyze digital inequalities and subsequently employs it to evaluate the effectiveness of “LEHAVA,” the largest government-sponsored program aimed at mitigating these disparities in Israel.

Optimizing Strategic Decision-Making in Multinational Corporations through Data Mining-Based User Behavior Analysis

Optimizing Strategic Decision-Making in Multinational Corporations through Data Mining-Based User Behavior Analysis

Customer Segmentation in User Behavior Analysis: A Comparative Study of Clustering Algorithms

A thorough understanding of customer behavior patterns is still essential for corporate success in today's digital environment. To segment clients in-depth, this study used three distinct clustering algorithms: k-means, hierarchical clustering, and dbscan. By carefully examining the age, yearly income, and consumption score This study goes beyond conventional views and creates a comprehensive and distinct perspective when disclosing numerous consumer attributes using data derived from the mall consumer Segmentation Data set. It offers insightful information that can be used to adjust a company's marketing plan. The strategic application of customer insights is redefined in this study, empowering stakeholders to decide wisely and improve market performance. The road to greater competitiveness and relevance in a developing market segment is mapped using real-world data and powerful clustering techniques. Based on research of this dataset, it was discovered that Dbscan performed best on this dataset with unequal density. It highlights the usefulness of these algorithms in today's complicated business world.

Outlier based intrusion detection in databases for user behaviour analysis using weighted sequential pattern mining

Outlier based intrusion detection in databases for user behaviour analysis using weighted sequential pattern mining

Do “Likes” in a Brand Community Always Make You Buy More?

Online brand communities often use social plug-in features, such as the Like button, to facilitate social interactions and engage users with the brands. However, whether and how such a community feature affects users’ purchases remain open questions. Analysis of user behavior following the adoption of the Like feature indicates a surprising downturn in purchases, with a 4.1% decrease in orders and a 25.0% reduction in expenditure. Notably, online purchases dip by 3.4% in order numbers and 21.1% in expenditure, with a slighter offline decrease. The treatment effect of the adoption is not always negative but varies over time and across users. First, the Like feature adoption has a positive effect on users’ purchases in the first two months (primarily through enhancing their community participation), and the treatment effect turns negative in subsequent months, leading to the overall negative treatment effect on purchases. Second, the negative treatment effect likely stems from unflattering social comparison and can become weaker or even positive when users accrue more Likes. However, only a small proportion of users receive sufficient Likes to be motivated to purchase more. Our results caution against potential downsides of the Like feature in online communities and provide valuable managerial implications.

Exploration of Effective Methodologies for Web Personalization

In the dynamic landscape of online interactions, the exploration of effective methodologies for web personalization emerges as a critical step in the digital marketing strategy of a brand. As users navigate through an increasingly vast digital realm, the ability to tailor online experiences becomes pivotal for engaging and retaining diverse audiences. This study delves into the multifaceted realm of web personalization, aiming to understand methodologies that prove efficacious in crafting tailored digital experiences. Key considerations include the strategic deployment of technology, the role of data analytics, and the implementation of a buyer persona to better analyze user behavior. By evaluating diverse approaches, this exploration seeks to unravel best practices that resonate with customers, foster heightened user engagement, and contribute to the evolving landscape of personalized online experiences. Ultimately, this research aims to provide insights that empower businesses and digital platforms to navigate the complexities of web personalization successfully.

NAVIGATING DISTRACTIONS: UNVEILING THE POWER OF DIGITAL ADVERTISING IN LOW-ATTENTION ENVIRONMENTS

Unveiling the Power of Digital Advertising in Low-Attention Environments," delves into the dynamics of digital advertising's effectiveness when faced with challenges posed by low-attention environments. In an era saturated with information and distractions, the study investigates strategies and elements within digital advertising that capture and retain audience focus. Through a comprehensive analysis of engagement metrics, user behaviors, and content attributes, the research aims to provide insights for advertisers seeking to optimize their campaigns in settings where attention is at a premium.

Usable Security: A Systematic Literature Review

Usable security involves designing security measures that accommodate users’ needs and behaviors. Balancing usability and security poses challenges: the more secure the systems, the less usable they will be. On the contrary, more usable systems will be less secure. Numerous studies have addressed this balance. These studies, spanning psychology and computer science/engineering, contribute diverse perspectives, necessitating a systematic review to understand strategies and findings in this area. This systematic literature review examined articles on usable security from 2005 to 2022. A total of 55 research studies were selected after evaluation. The studies have been broadly categorized into four main clusters, each addressing different aspects: (1) usability of authentication methods, (2) helping security developers improve usability, (3) design strategies for influencing user security behavior, and (4) formal models for usable security evaluation. Based on this review, we report that the field’s current state reveals a certain immaturity, with studies tending toward system comparisons rather than establishing robust design guidelines based on a thorough analysis of user behavior. A common theoretical and methodological background is one of the main areas for improvement in this area of research. Moreover, the absence of requirements for Usable security in almost all development contexts greatly discourages implementing good practices since the earlier stages of development.

Deep Learning Algorithms for Personalized Services and Enhanced User Experience in Libraries

The integration of deep learning (DL) algorithms in library settings engenders a multitude of challenges and complexities, encompassing unintended ramifications, ethical quandaries, a dearth of specialized literature elucidating DL in library contexts, the intricacies of dataset selection and human intervention, and the inherent limitations when juxtaposed with the remarkable cognitive capabilities of the human brain. To surmount these hurdles and attain a profound comprehension of DL in library settings, a rigorous and comprehensive systematic literature review (SLR) becomes imperative. This study investigates the application of DL algorithms in examining user-seeking behaviour to provide personalized services and enhance user experience in libraries. Through a comprehensive literature review, the study aims to uncover the benefits, challenges, and implications of integrating DL algorithms for user behaviour analysis and personalized services in library environments. The investigation encompasses a systematic literature review, employing a meticulous search and screening process utilizing the Scopus database. DL algorithms enable tailored recommendations, resource suggestions, and personalized search outcomes, improving information retrieval and user-centric services. Ethical considerations and ongoing research are emphasized to address challenges and maximize the potential of DL algorithms in libraries. The integration of DL algorithms in libraries yields substantial benefits, including improved information retrieval capabilities, augmented resource recommendation systems, and the delivery of user-centric services. The paper offers valuable insights to researchers, practitioners, and stakeholders operating within this field.

Electricity trading of multiple home microgrids through V2X based on game theory

Currently, a significant number of householders are contentedly participating in energy markets integrated with renewable energy sources (RES) to control global warming. Further, the high penetration of e-mobility influences the net zero concepts globally. However, uncoordinated scheduling of electric vehicles (EV) battery charging/ discharging parked in residential areas would result in critical issues in the power system in terms of energy availability. Thus, running intelligent charging/ discharging approaches can tackle these obstacles while managing green energy efficiently. This paper proposes a novel smart EV charging/ discharging strategy for multiple home microgrids (H-MG) based on user behavior analytics patterns with a non-cooperative game theory approach named Nikaido–Isoda/Relaxation algorithm (NIRA). In the proposed structure, EVs are active loads besides responsive loads in H-MGs and are involved in demand-side management (DSM) to achieve peak shaving. Furthermore, electricity retailers (ER) are encountered with distinct properties following opposite intentions and could participate in the competitive retail electricity market (REM) in residential areas. Therefore, households can maximize their profit by reducing varied costs, such as EV charging and battery degradation, while minimizing their carbon footprint. In addition, this study works on the distributed decision-making process that could provide optimal power-sharing and the best payoff in an unbiased procedure. After crucial investigations, the simulation outcomes demonstrate the effectiveness of the proposed approach in encouraging various households to participate actively in the REM with other agents to increase their profits. According to acquired results, active residents saved their budget 28%–45% through employing intelligent EV charging/ discharging strategies in the REM with dynamic pricing compared to the fixed tariff.

Topic Modelling of Web Pages with Latent Dirichlet Allocation Methods

Topic modelling with Latent Dirichlet Allocation (LDA) is a popular technique used in natural language processing to uncover hidden thematic structures within a collection of documents. When applied to web pages, LDA can help in identifying prevalent topics or themes across these pages.. This study delves into the utilization of Latent Dirichlet Allocation (LDA) methods to extract underlying topics within web pages, a fundamental pursuit in understanding the multifaceted landscape of online information. Web content analysis presents unique challenges owing to its diverse nature—comprising text, images, videos, and structured HTML elements—mandating rigorous preprocessing strategies to homogenize the data. By adapting the LDA model to accommodate these challenges, this research tackles the task of uncovering latent thematic structures prevalent across web content. Methodologically, the study explores parameter tuning and model adaptation to optimize LDA for web page analysis, navigating complexities such as varied content formats, noise, and inherent biases in web data. Addressing these intricacies involves parsing HTML, extracting meaningful textual information, and refining tokenization processes. Evaluating the fidelity and interpretability of discovered topics becomes pivotal, prompting the utilization of coherence scores, perplexity metrics, and human assessment to gauge the quality of generated topics. Additionally, this research confronts the dynamic nature of web content, proposing strategies like continuous model retraining and dynamic topic modeling to accommodate evolving trends and updates. Practical applications of the extracted topics span a spectrum of domains, encompassing content recommendation systems, user behavior analysis, sentiment analysis, targeted advertising, and the enhancement of search algorithms for improved relevance and user engagement. Supported by illustrative case studies, this study elucidates how LDA serves as a potent mechanism to distill coherent and meaningful topics from web pages, offering invaluable insights into the hidden structures within the vast expanse of online information. St This comprehensive abstract encapsulates the depth and breadth of employing LDA for the analysis of web content, encompassing challenges, methodologies, evaluations, applications, and real-world implications.

Maximize The Use of Internet To Improve Sales of Small Businesses: Evidence from Indonesia

In today's digital era, businesses must harness the internet’s full potential to improve their sales and remain competitive. Unfortunately, not all small businesses in Indonesia can improve their sales on internet. An official survey by APJII (Indonesia Internet Service Provider Association) found that estimates of sales growth of small businesses using the internet are not maximal, and some of them haven’t grown in sales. On the other side, case studies showcasing successful internet-driven sales technique implementations. Therefore, this study analyzes how small businesses can maximize internet use to optimize their sales performance and explores strategies and best practices. A comprehensive analysis of literature, secondary data, usage and user behavior of the internet in Indonesia was used. By understanding and applying the insights presented in this research, small businesses have the ability to take advantage of the internet and achieve considerable increases in their sales figures.

Sports Work Strategy of College Counselors Based on MySQL Database Big Data Analysis

In order to turn our attention to the real progress of the education world, this paper makes a concrete study on the efficiency optimization of MySQL database under big data. On the basis of discussing the significance of cultivating students' team spirit, this paper analyzes and discusses the strategies of cultivating students' team spirit in extracurricular sports activities. MySQL database, as one of the widely used open source databases, naturally has relatively excellent running efficiency, but it also poses a strong challenge for further improving the running efficiency of MySQL database. In many external services, databases are needed, whether in websites or other large-scale software. All the mobile users who need to use the database to store a large amount of user data have caused many websites to collect users and user behavior analysis, which can make them better optimize the website and products, so the dependence and application of data are increasing, and the optimization of databases can greatly improve the ability of developers to retrieve and analyze data.

Research and improvement of movie recommendation based on a collaborative filtering algorithm

Among various recommendation algorithms, collaborative filtering stands out as a popular and effective technique that analyzes user behavior to suggest personalized items. The goal of studying recommendation algorithms is to enhance the exactness of recommendation outcomes and provide users with tailored suggestions. In this paper, the algorithm of collaborative filtering is divided into User-CF and Item-CF, and their algorithm principles and implementation steps are introduced. The similarity calculation of both methods is improved, and a comparison of their advantages and disadvantages is conducted. Experiments are conducted on the MovieLens-latest-small movie dataset to compare the performance of User-CF and Item-CF before and after their similarity calculation is improved. The results show that the similarity improvement of User-CF can substantially enhance the algorithms inclusiveness, while the similarity improvement of Item-CF does not have a considerable impact on the algorithms performance indicators. Finally, the comparison shows that Item-CF has better coverage and popularity, while User-CF has better precision and recall.

Application of sensor technology in grasping and preprocessing of network hotspot information propagation

The rapid dissemination of hot information on the Internet has become a common phenomenon in today's society. Traditional methods of information capture and preprocessing often require a lot of manpower and material resources, and the captured information has low timeliness and accuracy. The purpose of this paper was to use sensor technology to find and locate network hotspots in time. By collecting user generated content, social media data, news reports, etc., the data is analyzed and mined to identify popular topics and events. In terms of information capture, sensor technology can monitor and understand user activities, the popularity of posts, emotional tendencies, user attention, user interaction, etc., through information monitoring. Network data was collected, such as network latency, data loss rate, and bandwidth utilization. Sensor technology was used to collect social media data to understand the level of public attention to hot events. In information preprocessing, sensor technology was used to remove noise and redundant information in data to ensure data quality. The data were labeled and classified, and the information dissemination rules of network hotspot were analyzed in depth. The average capture accuracy of Method 1 for Hotspot 1, Hotspot 2, and Hotspot 3 was 72.11%, 71.81%, and 72.54%, respectively. The average capture accuracy of Method 2 for Hotspot 1, Hotspot 2, and Hotspot 3 was 82.55%, 83.14%, and 82.91%, respectively. When the data was 40, 80, and 120, the preprocessing times of Method 1 for Post 1 were 8.81 s, 15.47 s, and 18.77 s, respectively. The preprocessing times of Method 2 for Post 1 were 5.97 s, 7.80 s, and 9.25 s, respectively. The application of sensor technology in the capture and preprocessing of network hot information dissemination has brought a variety of innovations, including multi-modal data acquisition, real-time monitoring and analysis, user behavior analysis, data cleaning and integration, anomaly detection and early warning, intelligent recommendation and personalized service, etc., thus improving the accuracy, real-time and personalized degree of information acquisition.

Racial Discrimination and Anti-discrimination: The COVID-19 Pandemic’s Impact on Chinese Restaurants in North America

The coronavirus disease 2019 (COVID-19) pandemic has seen a rise in racial discrimination against Asian communities, notably the Chinese population. Despite growing research on various aspects of the pandemic, there is a notable gap in understanding its behavioral impact regarding racial discrimination. This study delves into the manifestations of COVID-19-related racial discrimination and antidiscrimination efforts on online platforms using large-scale data sets from Yelp.com and SafeGraph. We specifically examined how the pandemic affected Chinese restaurants compared with non-Chinese ones at different pandemic phases. Our findings are significant; the pandemic triggered an immediate surge in racial discrimination, resulting in a substantial decrease in customers visiting Chinese restaurants. Importantly, we applied advanced text mining and machine learning techniques to analyze user behavior, consistently revealing that increased discrimination prompted users to take antidiscrimination actions on online platforms. This research highlights a tangible form of racial discrimination through reduced patronage of Chinese restaurants and underscores the capacity of consumers to combat discrimination on online platforms. It calls for targeted policy interventions to address and prevent racial discrimination, particularly in the context of public health crises.

Dual-channel early rumor detection based on factual evidence

Social media facilitates people's free expression and communication, but it also provides a platform for the generation and dissemination of rumors. In order to curb the spread of rumors in time and minimize their harm, early rumor detection methods were born and became a research hotspot. Early rumor detection requires a balance between accuracy and timeliness, but most existing methods only focused on accuracy and neglected efficiency. In this regard, we analyze a large number of Internet rumors and found that some rumors have been repeatedly spread after a small amount of modification. Such old rumors can be quickly and easily identified through factual evidence, while other new rumors require more clues and time to crack. Based on this insight, we propose a dual-channel early rumor detection model named ERD-DC (Early Rumor Detection-Dual-Channel), which leverages deep learning algorithms. Specifically, ERD-DC consists of two distinct channels, a fast channel referred to as channel-E (channel-Easy) and a slow channel named channel-D (channel-Difficult). In the ERD-DC, posts are divided into ‘easy-to-identify’ and ‘difficult-to-identify’ based on the availability of relevant evidence online. Subsequently, both types of posts are directed to their respective channels for detection. Channel-E focuses on identifying easy-to-identify rumors through a straightforward and swift post-evidence matching process, primarily aimed at enhancing timeliness. In parallel, channel-D is tasked with uncovering difficult-to-identify rumors, leveraging a more comprehensive analysis of user behaviors and posts transmission structures to elevate overall accuracy. Experimental results on two real datasets demonstrate that ERD-DC outperforms other state-of-the-art models significantly in early rumor detection with an accuracy rate of 84%.

Data adjusting strategy and optimized XGBoost algorithm for novel insider threat detection model

With the growth of access to internal system resources, the insider threat problem is emerging and can bring immeasurable losses to enterprises. In order to detect the hidden threats and guide the formulation of enterprise management strategy, it is important to analyze and understand the employee behavior. Thus, we suggest a user behavior analysis system framework, which mainly includes data processing, user behavior modeling and results analysis. Data Adjusting (DA) strategy and optimized eXtreme Gradient Boosting (XGBoost) model are utilized with the aim of full analysis small amount of feature information. The strategy for detecting suspicious behavior can be the following. Firstly, select initial suspicious data, misclassification data retention and combination sampling. Secondly, for further behavior model construction, an improved Particle Swarm Optimization algorithm based on ethnic randomized particles (ERPSO), which introduces Gaussian white noise with adjustable intensity into acceleration coefficients is given for searching the optimal XGBoost parameters. In addition, based on the designed DA strategy and the proposed ERPSO algorithm, we have also compared the results of the proposed methods with the current state-of-the-art methods. Experimental results show that the XGBoost optimized by the ERPSO (ERPSO-XGBoost) model has comprehensive performance, which proves the rationality and effectiveness of the insider behavior analysis system framework. Through a comprehensive understanding of insider behavior, the obvious characteristic behavior is found to adjust the management strategy and guide the behavior modeling, so as to prevent more losses in time.