I am a Senior Member of Technical Staff in the Database Manageability group at Oracle.

Prior to joining Oracle, I worked as a Health Systems Software Engineer at the Vanderbilt Institute for Clinical and Translational Research (VICTR). I was a member of the group that develops BioVU as well as the Research Derivative and Synthetic Derivative data warehouses.

Previously I was a Postdoctoral Researcher at the Department of Computer Science and Engineering of the University of California, Riverside as a member of the database lab directed by Associate Professor Vagelis Hristidis.

I received my PhD from the School of Electrical and Computer Engineering of the National Technical University of Athens, Greece (NTUA) under the supervision of Prof. Yannis Vassiliou and Prof. Timos Sellis. I was a member of the Knowledge and Database Systems Laboratory, NTUA and the Institute for the Management of Information Systems, R.C. Athena.

I received my Diploma in Electrical and Computer Engineering from the School of Electrical and Computer Engineering, NTUA in 2005.

My research interests include: Healthcare Analytics, Personalization in data management and information retrieval, Recommender systems, Spatial databases, Flexible query answering and Context-aware computing

Selected Publications

Theses

  • Anastasios Arvanitis, Personalized Data Management Systems, PhD Dissertation, Knowledge and Database Systems Laboratory, School of Electrical and Computer Engineering, National Technical University of Athens, Greece, January 2013
  • Anastasios Arvanitis, System and Application Integration using Web Services, Diploma Thesis, Knowledge and Database Systems Laboratory, School of Electrical and Computer Engineering, National Technical University of Athens, Greece, July 2005 (Diploma Thesis supervisor: Professor Timos Sellis)

Presentations and Talks

  • Efficient Influence-Based Processing of Market Research Queries, Hellenic Data Management Symposium (HDMS'14), Athens, Greece, July 24th, 2014
  • Efficient Concept-based Document Ranking, EDBT'14, Athens, Greece, March 27th, 2014
  • Topics in Personalized Data Management, Invited Talk, University of Southern California, Los Angeles, CA, USA, May 8th, 2013
  • Personalized Data Management Systems, PhD Defense, NTUA, Athens, Greece, January 4th, 2013
  • Efficient Influence-Based Processing of Market Research Queries, CIKM'12, Maui, HI, USA, November 1st, 2012
  • User-Centered Data Management, Invited Talk, University of California - Riverside, Riverside, CA, USA, May 29th, 2012
  • PrefDB: Supporting Preferences as First-Class Citizens in Relational Databases, Invited Talk, IMIS, Athens, Greece, May 17th, 2012
  • Towards Preference-aware Relational Databases, ICDE'12, Washington DC, USA, April 2nd, 2012
  • Personalized Data Management Systems, Thesis Proposal, NTUA, Athens, Greece, February 8th, 2011
  • Towards Preference-aware Relational Databases, Invited Talk, IMIS, Athens, Greece, November 26th, 2010
  • Probabilistic Contextual Skylines, Hellenic Data Management Symposium (HDMS'10), Ayia Napa, Cyprus, July 2nd, 2010
  • Probabilistic Contextual Skylines, ICDE'10, Long Beach, CA, USA, March 2nd, 2010
  • Probabilistic Contextual Skylines, DB Seminars Talk, NTUA, Athens, Greece, January 11th, 2010
  • Finding the top-k Influential Points through Reverse Skylines, HKUST, Hong Kong, China, May 20th, 2009
  • A Survey of Context and Context-aware Data Management, DB Seminars Tutorial, NTUA, Athens, Greece, December 10th, 2007

Research

During my work at Vanderbilt, I was involved in building and maintaining the data infrastructure of the Research Derivative and Synthetic Derivative data warehouses of the Vanderbilt University and Medical Center. My focus was on developing scalable data pipelines and applications for processing and analyzing structured and unstructured medical data using a variety of database, data mining and natural language processing techniques. RD and SD data warehouses maintain over 200M medical records for about 2.2M patients, with more than 100,000 new records being added every day. Generated data are used in various clinical studies conducted in Vanderbilt University and Medical Center, as well as in SPHINX research project by EMERGE network partners.

As part of my research I have worked in several projects including:

  • ConceptRank (joint work with M. Wiley and V. Hristidis) [EDBT'14]
    • Recently, there is increased interest in searching and computing the similarity between Electronic Medical Records (EMRs). Although several rich biomedical ontologies and sophisticated similarity metrics have been proposed, important scalability challenges arise for supporting keyword search and similarity queries over EMR databases. In this project we make a first step to address these challenges. In particular I was involved in developing top-k ranking techniques for medical record documents, which leverage the relationships and structure of domain ontologies. We also experimentally evaluated the performance of the proposed methods on a large publicly available EMR dataset.
  • PrefDB system (joint work with G. Koutrika) [ICDE'12], [SIGMOD'12], [TKDE]
    • Worked on designing and implementing the PrefDB prototype system. PrefDB acts as a personalization layer on top of a RDBMS and facilitates engineering and execution of applications that require preference-aware query processing (e.g. ranked query results). PrefDB is built on top of PostgreSQL and includes several components such as a profile manager, a query parser, optimizer and execution engine as well as an administration/monitoring tool. In PrefDB each user preference is evaluated as a new operator that assigns scores to affected tuples. The query optimizer allows cost-based query optimizations based on the preferences properties (e.g. selectivity), database statistics following a novel cost model and plan enumeration techniques.
    • PrefDBAdmin: A graphical tool that helps application designers in the process of building, testing and fine-tuning preferential queries on top of a relational DBMS.
  • Skyline Query Processing
    • Probabilistic Contextual Skylines. Developed a probabilistic adaptation of skyline query algorithms for both non-indexed and indexed (using R-tree indexes) data. This work relaxes the requirement from the users to explicitly state their preferences by extrapolating uncertain preferences based on the user's context (e.g. current location, time) and previous behavior. [ICDE'10]
    • Reverse Skyline Queries for Market Analysis Problems. Developed novel algorithms applied in personalized advertising and product positioning. The proposed algorithms outperform the state-of-the art approaches in terms of efficiency, scalability (scale to millions of product or customer data and multiple dimensions) and progressiveness of result calculation. [CIKM'12]

I have also worked as a Graduate Research Assistant at the Institute for the Management of Information Systems - R.C. Athena (IMIS) (December 2008-June 2010),

  • TALOS (project partially funded under FP7 - European Union)
    • The scope of this project was to deliver location-based services for mobile environments by (a) enriching content with task and spatiotemporal semantics, (b) developing a user interface and a prototype service that provides efficient discovery of relevant content based on the principles of task computing and by applying state-of-the-art approximate positioning techniques.

Academic Service

I haved served as a program commitee member for:

as a reviewer for:

as an external reviewer for:

as the Web/Information Chair for:

and as a Staff Member for:

Teaching

I have worked as a Teaching Assistant at the School of Electrical and Computer Engineering of the National Technical University of Athens (NTUA) on:

and as an assistant supervisor for the Diploma Theses of:

During my work at the University of California, Riverside I have collaborated with the following graduate students:

Work Experience

You can visit my LinkedIn profile for a detailed bio of my work experience

Link to my resume