Data Science Lab

Introduction

Behavior Informatics, and Behavioral AI, [1,2] refer to a research area that aims to develop methodologies, techniques and practical tools for representing, modeling, analyzing, understanding, managing and utilizing observable, symbolic, mapped and non-occurring behaviors, behavioral interactions, behavior networks, behavior patterns, behavior impact, the formation of behavior-oriented groups, collective intelligence, consequences, the emergence of behavioral intelligence, and their causes and evolution.

In essence, behavioral AI, analytics, computing and informatics seek to deliver quantitative and computational technologies and tools to deeply understand behaviors, social behavior networks, their evolution, effect and impact. In this sense, we also call it behavioral computing. As a research issue, BI consists of many research directions that are worthy of systematic research and conducting case studies from aspects such as behavioral data construction, behavior modeling and representation, behavior impact modeling, behavior pattern analysis, behavior network analysis, non-occurring behavior analysis [3], and behavior intervention and management. Additionally, behavior measurement and evaluation, behavior presentation, and behavior use are very important topics.

Why Behavioral AI, Computing and Informatics

Behavior forms a critical ingredient and engine of human beings, artificial systems, business, society, environment, and economy. Behavior is ubiquitous, and behavior intelligence drives the development, consequence, and evolution of our human beings, artificial systems, business, society, environment, and economy.

Complex behaviors are widely seen in artificial and natural intelligent systems, on the internet, physical and virtual systems, social and online networks, multi-agent systems, and brain systems. The in-depth understanding of complex behaviors has been increasingly recognized as a crucial means for disclosing interior driving forces, causes and impact on businesses in handling many challenging issues.

However, traditional behavior modeling mainly relies on qualitative methods from behavioral science and social science perspectives. The so-called behavior analysis in behavior science and data analytics and learning often focuses on human demographic and business usage data, in which behavior-oriented elements are hidden in routinely collected transactional data. As a result, it is ineffective or even impossible to deeply scrutinize native behavior intention, lifecycle, dynamics and impact on complex problems and business issues.

Behavioral AI, analytics, computing and informatics aim for an in-depth analysis of behaviors and their interior driving forces, causes and impact. Behavior informatics consists of theoretical and applied studies on high impact behavior sequence analysis, impact-oriented combined behavior analysis, high utility behavior analysis, nonoccurring behavior analysis, coupled, group and collective behavior analysis [4], statistical modeling of coupled behaviors, probabilistic modeling of sparse rating behaviors, understanding behavior choice and attraction, behavior analysis with recurrent networks, behavior analysis in visual data, behavior learning from demonstrations.

Research Topics

The research topics include but are not limited to the following areas:

What is Behavior: abstract behavior model to capture both intrinsic and contextual properties of behaviors from both subjective and objective perspectives;
High Impact and High Utility Behavior Analysis: algorithms and case studies to identify behaviors associated with high impact and utility;
Nonoccurring Behavior Analysis: algorithms and case studies about analyzing and mining negative behavior sequences in an efficient way;
Coupled, Group and Collective Behavior Analysis: algorithms and case studies for identifying and analyzing group/community behaviors and anomalies;
Statistical Modeling of Coupled Behaviors: statistical models to induce explicit and implicit couplings between attributes into statistical models for modeling behaviors;
Understanding Behavior Intent: choice modeling, attraction learning and deep models to learn user choice and attraction and sequential interactions;
Behavior Analysis with Recurrent Networks: deep recurrent networks to address behavior sequence analysis problems using several representative cases;
Behavior Analysis in Visual Data: traditional approaches and recently emerged deep learning-based approaches for learning visual behaviors;
Behavior Learning from Demonstrations: advanced imitation learning approaches to imitate human behaviors in a set of demonstrations data;
Challenges and Prospects: open issues and potential of complex behavior modeling, analysis and mining in terms of both qualitative and quantitative aspects.

The following diagram shows a concept map of Behavior Informatics.

Figure 1. The research map of behavior informatics [1].

Related Tutorials

Longbing Cao and Can Wang. Behavior Informatics: Methods and Applications at AAMAS2020.
Longbing Cao. Behavior Analytics: Methods and Applications, AAAI2019 tutorial, slides.
Longbing Cao, Philip Yu, Guansong Pang. KDD2018 tutorial: Behavior analytics: Methods and applications.
Longbing Cao, Behavior Computing: Deep Behavior Analytics and Active Behavior Management, PAKDD2015, Ho Chi Mingh City, Viet Nam.
Longbing Cao, Philip S Yu, Can Wang. IJCAI2013 tutorial: Behavior informatics – Modeling, analysis and mining of complex behaviors
Longbing Cao, Philip S Yu, Can Wang Behavior Computing: Complex Behavior Modeling, Analysis and Mining, WI-IAT 2012, 4 Dec 2012, Macau, China.

References

	Longbing Cao, Philip S Yu (Eds). Behavior Computing: Modeling, Analysis, Mining and Decision, ISBN: 978-1-4471-2968-4, Springer, 2012. BibTeX
	Longbing Cao; Hiroshi Motoda, Jaideep Srivastava, Ee-Peng Lim, Irwin King, Philip S. Yu, Wolfgang Nejdl, Guandong Xu, Gang Li, Ya Zhang (Eds.). Behavior and Social Computing, Proceedings of International Workshop on Behavior and Social Informatics and Computing, Lecture Notes in Computer Science, Vol. 8178, Springer, 2013

[1] Longbing Cao, In-depth Behavior Understanding and Use: the Behavior Informatics Approach, Information Science, 180(17); 3067-3085, 2010. BibTeX.
[2] Longbing Cao, et al. Behavior Informatics: A New Perspective. IEEE Intelligent Systems (Trends and Controversies), 29(4): 62-80, 2014. BibTeX.
[3] Longbing Cao, Philip S. Yu, Vipin Kumar. Nonoccurring Behavior Analytics: A New Area. IEEE Intelligent Systems 30(6): 4-11 (2015). BibTeX.
[4] Longbing Cao, Yuming Ou, Philip S Yu. Coupled Behavior Analysis with Applications, IEEE Trans. on Knowledge and Data Engineering, 24(8): 1378-1392 (2012). BibTeX.
Longbing Cao. Behavior Informatics to Discover Behavior Insight for Active and Tailored Client Management. KDD2017, 2017. BibTeX.
Herath, S., Harandi, M., Porikli, F. (2017). Going deeper into action recognition: A survey. Image and vision computing, 60, 4-21.
Hussein, A., Gaber, M. M., Elyan, E., Jayne, C. (2017). Imitation learning: A survey of learning methods. ACM Computing Surveys (CSUR), 50(2), 21.
Longbing Cao, Xiangjun Dong and Zhigang Zheng. e-NSP: Efficient Negative Sequential Pattern Mining, Artificial Intelligence, 235: 156-182, http://dx.doi.org/10.1016/j.artint.2016.03.001, 2016. Download e-NSP codes and example; Download NegSeq codes and example BibTeX.
Zhigang Zheng, Wei Wei, Chunming Liu, Wei Cao, Longbing Cao, Maninder Bhatia. An effective contrast sequential pattern mining approach to taxpayer behavior analysis, World Wide Web 19(4): 633-651 (2016). BibTeX.
Longbing Cao. Coupling Learning of Complex Interactions, Journal of Information Processing and Management, 51(2): 167-186 (2015). BibTeX.
Jingyu Shao, Junfu Yin, Wei Liu,, Longbing Cao. Mining actionable combined patterns of high utility and frequency. DSAA 2015: 1-10, Research Track. BibTeX.
Can Wang,Longbing Cao, Chi-Hung Chi. Formalization and Verification of Group Behavior Interactions. IEEE T. Systems, Man, and Cybernetics: Systems 45(8): 1109-1124 (2015). BibTeX.
Longbing Cao. Combined Mining: Analyzing Object and Pattern Relations for Discovering and Constructing Complex but Actionable Patterns, WIREs Data Mining and Knowledge Discovery, 3(2): 140-155, 2013. BibTeX.
Junfu Yin, Zhigang Zheng, Longbing Cao, Yin Song, Wei Wei. Efficiently Mining Top-K High Utility Sequential Patterns, ICDM2013: 1259-1264. BibTeX.
Junfu Yin, Zhigang Zheng, Longbing Cao. USpan: An Efficient Algorithm for Mining High Utility Sequential Patterns, KDD 2012, 660-668. BibTeX.
Song, Y., Cao, L. et al. Coupled Behavior Analysis for Capturing Coupling Relationships in Group-based Market Manipulations. KDD 2012, 976-984, 2012.
in Song and Longbing Cao. Graph-based Coupled Behavior Analysis: A Case Study on Detecting Collaborative Manipulations in Stock Markets, IJCNN 2012, 1-8. BibTeX.
Wang, H., Klser, A., Schmid, C., Liu, C. L. (2011). Action recognition by dense trajectories. In CVPR (pp.3169-3176).
Longbing Cao, Yuming Ou, Philip S YU, Gang Wei. Detecting Abnormal Coupled Sequences and Sequence Changes in Group-based Manipulative Trading Behaviors, KDD2010, 85-94. BibTeX.
Zhigang Zheng, Yanchang Zhao, Ziye ZuoLongbing Cao, Huaifeng Zhang, Yanchang Zhao, Chengqi Zhang. An Efficient GA-Based Algorithm for Mining Negative Sequential Patterns, PAKDD2010, 262-273. BibTeX.
Yanchang Zhao, Huaifeng Zhang, Shanshan Wu, Jian Pei,Longbing Cao, Chengqi Zhang and Hans Bohlscheid. Debt Detection in Social Security by Sequence Classification Using Both Positive and Negative Patterns, ECML/PKDD2009, 648-663, 2009. BibTeX.
Yanchang Zhao, Huaifeng Zhang, Longbing Cao, Chengqi Zhang and Hans Bohlscheid. Mining Both Positive and Negative Impact-Oriented Sequential Rules From Transactional Data, PAKDD2009, pp.656-663. BibTeX.
Yeffet, L., Lior W. (2009). Local trinary patterns for human action recognition. In CVPR (pp. 492-497).
Longbing Cao. Zhao Y., Zhang, C. Mining Impact-Targeted Activity Patterns in Imbalanced Data, IEEE Trans. on Knowledge and Data Engineering, 20(8): 1053-1066, 2008. BibTeX.
Longbing Cao, Yanchang Zhao, Chengqi Zhang, Huaifeng Zhang. Activity Mining: From Activities to Actions, International Journal of Information Technology & Decision Making, 7(2): 259-273, 2008 BibTeX.
Laptev, I., Marszalek, M., Schmid, C., Rozenfeld, B. (2008). Learning realistic human actions from movies. In CVPR (pp. 1-8).

Notes:
For references on Nonoccurring Behavior Analytics and Negative Sequence Analysis, please refer to Nonoccurring Behavior and Negative Sequence Analysis for more information.
Please also refer to the Behavior Informatics website for more information about the relevant activities and publications.