A System Framework for Personalized and Transparent Data-Driven Decisions

Decision support systems that rely on data analytics are used in numerous applications. Their advantages are indisputable, however, they also present risks, possibly having severe impact on people’s lives. Consequently, the need to support ethical or responsible behavior of such systems has recently emerged, putting an emphasis on ensuring fairness, transparency, accountability, etc. This paper presents a novel system framework that offers transparent and personalized services tailored to user profiles to serve their best interest. Our framework personalizes the choice of model for individuals or groups of users based on metadata about data sets and machine learning models. Querying and processing these metadata ensures transparency by supporting various kinds of queries by different stakeholders. We discuss our framework in detail, show why existing solutions are inadequate, and highlight research questions that need to be tackled in the future. Based on a prototypical implementation, we showcase that even a baseline implementation of our framework supports the desired transparency and personalization.

[1]  Marek Kurzynski,et al.  A probabilistic model of classifier competence for dynamic ensemble selection , 2011, Pattern Recognit..

[2]  J. Reidenberg,et al.  Accountable Algorithms , 2016 .

[3]  Jenna Burrell,et al.  How the machine ‘thinks’: Understanding opacity in machine learning algorithms , 2016 .

[4]  Nojun Kwak,et al.  Feature extraction for classification problems and its application to face recognition , 2008, Pattern Recognit..

[5]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[6]  Ahmed Hosny,et al.  The Dataset Nutrition Label: A Framework To Drive Higher Data Quality Standards , 2018, Data Protection and Privacy.

[7]  Carlo Batini,et al.  Data Quality: Concepts, Methodologies and Techniques , 2006, Data-Centric Systems and Applications.

[8]  Carlo Batini,et al.  Data Quality: Concepts, Methodologies and Techniques (Data-Centric Systems and Applications) , 2006 .

[9]  Robert Sabourin,et al.  From dynamic classifier selection to dynamic ensemble selection , 2008, Pattern Recognit..

[10]  Robert Sabourin,et al.  A dynamic overproduce-and-choose strategy for the selection of classifier ensembles , 2008, Pattern Recognit..

[11]  Felix Naumann,et al.  Profiling relational data: a survey , 2015, The VLDB Journal.

[12]  Inioluwa Deborah Raji,et al.  Model Cards for Model Reporting , 2018, FAT.

[13]  Andrew D. Selbst,et al.  Big Data's Disparate Impact , 2016 .

[14]  Melanie Herschel,et al.  A survey on provenance: What for? What form? What from? , 2017, The VLDB Journal.

[15]  Timnit Gebru,et al.  Datasheets for datasets , 2018, Commun. ACM.

[16]  Kevin W. Bowyer,et al.  Combination of multiple classifiers using local accuracy estimates , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[17]  F Lefford,et al.  A blot on the profession , 1988, British medical journal.

[18]  Melanie Herschel,et al.  LuPe: A System for Personalized and Transparent Data-driven Decisions , 2019, CIKM.

[19]  Jeffrey O. Kephart,et al.  An artificial intelligence perspective on autonomic computing policies , 2004, Proceedings. Fifth IEEE International Workshop on Policies for Distributed Systems and Networks, 2004. POLICY 2004..

[20]  Abolfazl Asudeh,et al.  A Nutritional Label for Rankings , 2018, SIGMOD Conference.

[21]  Jianfeng Gao,et al.  A Diversity-Promoting Objective Function for Neural Conversation Models , 2015, NAACL.

[22]  Latanya Sweeney,et al.  Discrimination in online ad delivery , 2013, CACM.

[23]  Josep Domingo-Ferrer,et al.  Direct and Indirect Discrimination Prevention Methods , 2013, Discrimination and Privacy in the Information Society.

[24]  Emilio Corchado,et al.  A survey of multiple classifier systems as hybrid systems , 2014, Inf. Fusion.

[25]  R. Polikar,et al.  Ensemble based systems in decision making , 2006, IEEE Circuits and Systems Magazine.

[26]  Adam Tauman Kalai,et al.  Decoupled Classifiers for Group-Fair and Efficient Machine Learning , 2017, FAT.

[27]  Brian Kulis,et al.  Metric Learning: A Survey , 2013, Found. Trends Mach. Learn..

[28]  Toon Calders,et al.  Three naive Bayes approaches for discrimination-free classification , 2010, Data Mining and Knowledge Discovery.