A hierarchical multi-agent architecture based on virtual identities to explain black-box personalization policies

Abstract Hyper-personalization policies entail a considerable improvement regarding previous personalization approaches. However, they present several issues that need to be addressed, such as minimal explainability and privacy invasion. A hierarchical Multi-Agent System (MAS) is presented in this work to provide a solution to these concerns. The system is formulated as a hybrid approach, where some of the agents work autonomously, while the user input triggers the remaining. At the autonomous level, a set of Virtual Identities (VIs) representing different user profiles interact with Black-Box Hyper-Personalization Online Systems (BBHOS), gathering a set of targeted responses. Associative patterns and profile aggregations can then be inferred from the analysis of these responses. In the user-triggered level, the real user is virtualized as an identity that represents their features. The virtual identity serves as an intermediary between the personalization system and the real user. This virtualization hinders the personalization service from extracting sensitive contextual information about the real user, protecting their privacy. The results obtained by the user identity on its interaction with the personalization service are then analyzed, adjusting the content of the response to fit the user’s requests instead of their features. A use case on the functioning of the analysis of search engines is presented to illustrate the complete behavior of the proposed architecture.

[1]  Seung-A. Annie Jin,et al.  The virtual malleable self and the virtual identity discrepancy model: Investigative frameworks for virtual possible selves and others in avatar-based identity construction and social interaction , 2012, Comput. Hum. Behav..

[2]  Carlos Angel Iglesias,et al.  Validating viral marketing strategies in Twitter via agent-based social simulation , 2016, Expert Syst. Appl..

[3]  Emilio Serrano,et al.  An Intelligent and Autoadaptive System of Virtual Identities Based on Deep Learning for the Analysis of Online Advertising Networks , 2019, PAAMS.

[4]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Emad Abd-Elrahman,et al.  Performance Evaluation of Virtual Identity Approaches for Anonymous Communication in Distributed Environments , 2017, ANT/SEIT.

[6]  Xuefeng Zheng,et al.  A survey of semantic similarity and its application to social network analysis , 2015, 2015 IEEE International Conference on Big Data (Big Data).

[7]  Aren Jansen,et al.  CNN architectures for large-scale audio classification , 2016, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  Lefei Li,et al.  Personalization in Dynamic Assortment Planning: An Analysis Based on Multi-Agent Simulation Method , 2017, 2017 5th International Conference on Enterprise Systems (ES).

[9]  Sudeshna Sarkar,et al.  Ontology Based User Modeling for Personalized Information Access , 2010, Int. J. Comput. Sci. Appl..

[10]  Xiangjun Dong,et al.  A Context-Aware Mobile User Behavior-Based Neighbor Finding Approach for Preference Profile Construction , 2016, Sensors.

[11]  Stéphane Lecoeuche,et al.  Purchase intention-based agent for customer behaviours , 2020, Inf. Sci..

[12]  Wided Lejouad Chaari,et al.  Enhancing assessment of Personalized Multi-Agent System through ConvLSTM , 2017, KES.

[13]  Ting Liu,et al.  Document Modeling with Gated Recurrent Neural Network for Sentiment Classification , 2015, EMNLP.

[14]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[15]  Ramesh Nallapati,et al.  TopicFlow Model: Unsupervised Learning of Topic-specific Influences of Hyperlinked Documents , 2011, AISTATS.

[16]  Christopher E. Moody,et al.  Mixing Dirichlet Topic Models and Word Embeddings to Make lda2vec , 2016, ArXiv.

[17]  Jee Hyun Kim,et al.  A Context-Awareness Modeling User Profile Construction Method for Personalized Information Retrieval System , 2014, Int. J. Fuzzy Log. Intell. Syst..

[18]  Mahmood Yousefi-Azar,et al.  Text summarization using unsupervised deep learning , 2017, Expert Syst. Appl..

[19]  Abdelkader Ouda,et al.  Designing Security User Profiles via Anomaly Detection for User Authentication , 2020, 2020 International Symposium on Networks, Computers and Communications (ISNCC).

[20]  Tomas Mikolov,et al.  Bag of Tricks for Efficient Text Classification , 2016, EACL.

[21]  Wei Xu,et al.  CNN-RNN: A Unified Framework for Multi-label Image Classification , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Khalid Alfalqi,et al.  A Survey of Topic Modeling in Text Mining , 2015 .

[23]  Tao Zhang,et al.  Agent-based simulation of consumer purchase decision-making and the decoy effect , 2007 .

[24]  Geoffrey E. Hinton,et al.  Speech recognition with deep recurrent neural networks , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[25]  Ji-Rong Wen,et al.  Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation , 2021, WSDM.

[26]  Cheng-Yen Chang,et al.  Construct a Customized Product Service System Utilizing Multi-agent System , 2018, APMS.

[27]  Abdelaziz El Fazziki,et al.  Multi-agent framework for social CRM: Extracting and analyzing opinions , 2016, 2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA).

[28]  Przemyslaw Kazienko,et al.  AdROSA - Adaptive personalization of web advertising , 2007, Inf. Sci..

[29]  Emad Abd-Elrahman,et al.  A Novel Virtual Identity Implementation for Anonymous Communication in Cloud Environments , 2015, EUSPN/ICTH.

[30]  Feng Li,et al.  Offensive pricing strategies for online platforms , 2019, International Journal of Production Economics.

[31]  Tara N. Sainath,et al.  State-of-the-Art Speech Recognition with Sequence-to-Sequence Models , 2017, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[32]  Marcin Mironczuk,et al.  A recent overview of the state-of-the-art elements of text classification , 2018, Expert Syst. Appl..

[33]  Sung-Bae Cho,et al.  Honey Bee Behavior: A Multi-agent Approach for Multiple Campaigns Assignment Problem , 2008, 2008 International Conference on Information Technology.

[34]  Michalis Vazirgiannis,et al.  A Review of Web Document Clustering Approaches , 2010, Data Mining and Knowledge Discovery Handbook.

[35]  Xiangjun Dong,et al.  Correction: Gao, Q. et al. A Context-Aware Mobile User Behavior-Based Neighbor Finding Approach for Preference Profile Construction. Sensors 2016, 16, 143 , 2016, Sensors.

[36]  Aarti Singh,et al.  A Multi-agent Framework for Context-Aware Dynamic User Profiling for Web Personalization , 2019 .

[37]  Mohd Sharifuddin Ahmad,et al.  A Conceptual Multi-agent Semantic Web Model of a self-adaptive website for intelligent strategic marketing in learning institutions , 2016, 2016 2nd International Symposium on Agent, Multi-Agent Systems and Robotics (ISAMSR).

[38]  Juan A. Botía Blaya,et al.  Simulation of human behaviours for the validation of Ambient Intelligence services: A methodological approach , 2012, J. Ambient Intell. Smart Environ..

[39]  Zafeiris Kokkinogenis,et al.  Towards a Framework for Agent-Based Simulation of User Behaviour in E-Commerce Context , 2017, PAAMS.

[40]  Jacques Ferber,et al.  Multi-agent systems - an introduction to distributed artificial intelligence , 1999 .

[41]  Alex A. Freitas,et al.  Document Clustering and Text Summarization , 2000 .

[42]  Zenghui Wang,et al.  Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review , 2017, Neural Computation.

[43]  Anca L. Ralescu,et al.  Similarity Measure for Social Networks - A Brief Survey , 2015, MAICS.