I have been developing software for more than 15 years, and been doing it for a living for more than 9 year.
As a result, I was part of many projects, I try to list most of them here. My interests span a wide range of the data lifecycle, from management, to knowledge extraction and discovery, to information filtering and visualisation, as well as the
underlying software, AI principles, and the new computing infrastructures that make it all possible.
-
Follow-up: Analysis of Publications and Patents
Knowledge Discovery: Text Mining
2016-2018
I am part of a multi-year European Commission project on the Analysis of Publications and Patents of EU-founded projects, providing data management and extraction support, statistical and benchmarking insights, as well as a temporal and geographical pattern analysis on the evolution of keywords and key phrases over time and across countries and industries. -
BigAlgo
Knowledge Discovery: Data Science
2015-2016
Developing a software-as-a-service and platform-as-a-service solution for automated machine learning, data analysis, and business intelligence. -
Analysis of Publications and Patents
Knowledge Discovery: Text Mining
2012-2014
I am part of a multi-year European Commission project on the Analysis of Publications and Patents of EU-founded projects, providing data management and extraction support, statistical and benchmarking insights, as well as a temporal and geographical pattern analysis on the evolution of keywords and key phrases over time and across countries and industries. -
Collaborative Publication Disambiguation
Knowledge Extraction: Data Retrieval & Search
2012-2014
A federated search tool drawing data from various third-party APIs and SOAPs, and combining machine learning with a user-centric voting scheme, in order to learn and automate the proper matching, correction, and identification of publication references from free-form data. -
Statistical & Business Intelligence Dashboards
Exploratory Data Analysis & Visualisation
2007-2013
I've built various dashboards over the years, for monitoring business intelligence metrics, with dynamic aggregations by type, time range, department, etc. Many of them include features for status monitoring or reporting. -
Bespoke Content Management Systems
Data Management & Warehousing
2006-2013
I have developed various bespoke CMS (content management system) solutions over the years, aimed to support the business operations of my clients, by helping them digitize their processes. Many of them include product lifecycle management, product traceability, document templating, version comparison, print queues, user management, email functionality, or live search capabilities. -
A Hybrid Recommender System for Diverse Educational Programmes
Information Filtering & Recommender Systems
2012-2013
Recommender systems are everywhere in the digital world, shaping the choices we make, the products we buy, the books we read, or movies we watch. Few recommender systems exist, however, for the academic domain, where they could help students choose courses or academic programs. This thesis focuses on building a recommender system for this academic context, capable of supporting the needs of various types of academic departments. The main recommendation methods are analysed: content-based filtering, collaborative filtering, and demographic filtering; and different hybridization techniques are described. The idea of a single unifying model, a hybrid system with a statistical model that incorporates knowledge about items, users, and ratings, simultaneously, is introduced. The unifying hybrid framework is built and evaluated against the non-hybrid systems, using actual academic data from two different types of educational departments. -
Cross-Faculty Student Management and Data-Driven Recommendations
Data Management & Warehousing
2012
Part of a wider initiative to improve student and alumni relations, this project investigated the practical challenges and benefits of a cross-faculty data management solution for the student and alumni data of Maastricht University, as well as the challenges and benefits of a university-wide data analysis platform for information filtering and recommendations supporting alumni, students, educators, and the management staff. -
Interdisciplinarity of Nano Research Fields
Knowledge Discovery: Text Mining
2012
I developed an application for extracting the underlying controlled vocabularies of different knowledge thesauri and analysing the vocabulary data from a rough set perspective, computing the lower and upper approximations of keyword sets in order to uncover the Interdisciplinarity of Nano Research Fields. -
Workflow Adaptations
Data Management & Warehousing
2012
Helped my customers adapt various tools to fit their manufacturing workflow, by finding the right balance between using already available time-tested solutions and adding new custom functionality. -
University Data Integration Tool
Data Management & Warehousing
2011-2012
Data integration solution using ETL (extract, transform, load), extracting data from various data sources, e.g. SAP, Pandia, MS Excel, MS Word, CSV files, transforming and merging the data, and loading the combined and standardized data in new databases, e.g. MySQL, SQLite, or processing for display. -
Mining Student and Curriculum Data
Knowledge Discovery: Data Mining
2011-2012
This project investigated students' behavior and their academic evolution over time, as well as how these relate to student dropout, study length, and learning curves, for the purpose of developing valid student models. -
Master Orientation Tool
Information Filtering & Recommender Systems
2011-2012
I developed the recommendation engine that is part of a Master orientation tool – a curriculum planning application with exploration functionality for discovering future educational programs for students; the tool features a collaborative recommender system that matches the academic profiles of students with Master programs around the world. -
Data Mining Related to Determinants of Successful Educational Careers of Students
Knowledge Discovery: Data Mining
2011
The main objective of this project was to investigate the possibilities of applying Knowledge Discovery in Databases and Data Mining principles to the analysis of student data, aiming at finding patterns that relate meaningfully to student progress and success. Although Educational Data Mining (EDM) and Learning Analytics (LA) are growing more and more and are becoming recognized as fields of study in themselves, at the time of this project few research initiatives in these fields were focused on administrative data - as opposed to log dump files from interactive educational software, on building external models of students - as opposed to internal student modeling and knowledge tracing, or on outside-the-course or outside-curricula recommendations - as opposed to inside-the-course recommendations. This is where this project comes in, as it focused primarily on these under-studied topics. -
Predicting Stock Market Reactions to Corporate Press Releases
Knowledge Discovery: Text Mining
2011
Text Mining helps us process vast amounts of textual data in a fast manner using computers, in order to extract hidden knowledge within. One typical application is to use Natural Language Processing to classify documents based on their similarity to each-other. Eureka! If you would want to predict how something like, let's say the Stock Market, would react to something like, let's say a company press release, you would identify similar past releases with the current one, and bet your money on the fact that the market would react the same now. -
Twitter Happy Map
Knowledge Extraction: Information Retrieval
2011
Assessing the happiness level around the world, based on twitter messages moods. -
Knowledge Discovery in Student Data
Knowledge Discovery: Data Mining
2010
Applying Knowledge Discovery in Databases (KDD) and Data Mining (DM) principles to the analysis of the student data of university alumni. The main aim was to investigate whether this approach can bring novel insight into the study data, beyond classical statistical methods. -
Worth the switch?
Exploratory Data Analysis & Visualisation
2010
As there are some patents pending and coming up from various big companies about the topic covered here, I'm going to wait a bit and see how it all pans out before deciding how to continue with this topic. -
Non-choreographed Robot Dance
Artificial Intelligence
2009-2010
Teaching the Nao robot to dance on music. This research has tried to move away from the current choreographed approaches to Nao dance, and investigate how to make the robot dance in a non-predefined fashion. Aspects of both music and dance are investigated, and descriptions of how such elements were implemented in practice are presented. Moreover, focus has been put not only on the analysis of music's beat in order to build a dance, but other elements of music as well. -
Bug Tracking Migration Tool
Data Management & Warehousing
2009
Developed ETL (extract, transform, load) tools to move bug tracking data, by extracting from vendor-specific data structures, e.g. Bugzilla, transforming the data, and loading in other bug tracking databases, e.g. MantisBT. -
Privacy in a 2.0 World
Data Management & Warehousing
2009
A workshop about privacy that I conducted. -
ERP Migration Tool
Data Management & Warehousing
2008
Developed various data migration solutions, helping clients keep access to old business data when changing software vendors, by reverse engineering of legacy databases. I developed extract-transform-load tools to move enterprise resource planning data, by extracting data from legacy systems, e.g. FEPA, transforming the data structure, and loading in new ERPs, e.g. proALPHA. -
Event Management Platform
Data Management & Warehousing
2007
Helped develop an event management website, with membership management, social networking and email functionality, forum, listings, etc. -
Photo Sharing CDN
Cloud Computing
2006
Developed a geographically-distributed and provider-agnostic CDN (content delivery network) infrastructure and load balancing functionality for a photo sharing social network. -
Structure Builder
Data Management & Warehousing
2005
Developed a tool for generating a dynamic folder and file structure based on customer input, including naming patterns, templating of various document formats, automatic data insertion and form filling, etc.