ILIDViz:An incremental learning-based visual analysis system for network anomaly detection

Background With the development of information technology,there is a significant increase in the number of network traffic logs mixed with various types of cyberattacks.Traditional intrusion detection systems(IDSs)are limited in detecting new inconstant patterns and identifying malicious traffic traces in real time.Therefore,there is an urgent need to implement more effective intrusion detection technologies to protect computer security.Methods In this study,we designed a hybrid IDS by combining our incremental learning model(KANSOINN)and active learning to learn new log patterns and detect various network anomalies in real time.Conclusions Experimental results on the NSLKDD dataset showed that KAN-SOINN can be continuously improved and effectively detect malicious logs.Meanwhile,comparative experiments proved that using a hybrid query strategy in active learning can improve the model learning efficiency.

DFDVis:A Visual Analytics System for Understanding the Semantics of Data Flow Diagram

Data flow diagram(DFD),as a special kind of data,is a design artifact in both requirement analysis and structured analysis in software development.However,rigorous analysis of DFD requires a formal semantics.Formal representation of DFD and its formal semantics will help to reduce inconsistencies and confusion.The logical structure of DFD can be described using formalism of Calculus of Communicating System(CCS).With a finite number of states based on CCS,state space methods will help a lot in analysis and verification of the behavior of the systems.But the number of states of even a relatively small system is often very great that is called state explosion.In this paper,we present a visual system which combines Formal methods and visualization techniques so as to help the researchers to understand and analyze the system described by the DFD regardless of the problem of state explosion.

MBTIviz: A Visualization System for Research on Psycho-Demographics and Personality

The increasing interest in exploring the correlation between personal-ity traits and real-life individual characteristics has been driven by the growing popularity of the Myers–Briggs Type Indicator(MBTI)on social media plat-forms.To investigate this correlation,we conduct an analysis on a Myers–Briggs Type Indicator(MBTI)-demographic dataset and present MBTIviz,a visualiza-tion system that enables researchers to conduct a comprehensive and accessible analysis of the correlation between personality and demographic variables such as occupation and nationality.While humanities and computer disciplines provide valuable insights into the behavior of small groups and data analysis,analysing demographic data with personality information poses challenges due to the com-plexity of big data.Additionally,the correlation analysis table commonly used in the humanities does not offer an intuitive representation when examining the relationship between variables.To address these issues,our system provides an integrated view of statistical data that presents all demographic information in a single visual format and a more informative and visually appealing approach to presenting correlation data,facilitating further exploration of the linkages between personality traits and real-life individual characteristics.It also includes machine learning predictive views that help nonexpert users understand their personality traits and provide career predictions based on demographic data.In this paper,we utilize the MBTIviz system to analyse the MBTI-demographic dataset,calcu-lating age,gender,and occupation percentages for each MBTI and studying the correlation between MBTI,occupation,and nationality.

Metro-Wordle:An Interactive Visualization for Urban Text Distributions Based on Wordle

With the development of cities and the explosion of infonnation,vast amounts of geo-tagged textural data about Points of Interests(POIs)have been generated.Extracting useful information and discovering text spatial distributions from the data are challenging and meaningful.Also,the huge numbers of POIs in modem cities make it important to have efficient approaches to retrieve and choose a destination.This paper provides a visual design combing metro map and wordles to meet the needs.In this visualization,metro lines serve as the divider lines splitting the city into several subareas and the boundaries to constrain wordles within each subarea.The wordles are generated from keywords extracted from the text about POIs(including reviews,descriptions,etc.)and embedded into the subareas based on their geographical locations.By generating intuitive results and providing an interactive visualization to support exploring text distribution patterns,our strategy can guide the users to explore urban spatial characteristics and retrieve a location efficiently.Finally,we implement a visual analysis of the restaurants data in Shanghai,China as a case study to evaluate our strategy.