I am interested in intelligent web and social media systems with the focus on information retrieval, data summarization, recommendation systems, machine intelligence and data mining. During my PhD project, I developed several web system prototypes for enhanced exploration, browsing and search of social media data (news articles, blogs or social network content e.g., tweets from Twitter). I supervised several semester projects of master students with focus on web and mobile systems development. I have been a teaching assistant for Machine intelligence course (2 semesters). Below you can find more about my projects, research publications and presentations.
The aim of this project is to develop a novel system - a proof of concept that will enable more effective search, exploration, analysis and browsing of social media data. The main novelty of the system is an ad-hoc multi-dimensional topic map. The ad-hoc topic map can be generated and visualized according to multiple predefined dimensions e.g., recency, relevance, popularity or location based dimension. These dimensions will provide a better means for enhanced browsing, understanding and navigating to related relevant topics from underlying social media data. The ad-hoc aspect of the topic map allows user-guided exploration and browsing of the underlying social media topics space. It enables the user to explore and navigate the topic space through user-chosen dimensions and ad-hoc user-defined queries. Similarly, as in standard search engines, we consider the possibility of freely defined ad-hoc queries to generate a topic map as a possible paradigm for social media data exploration, navigation and browsing. An additional benefit of the novel system is an enhanced query expansion to allow users narrow their difficult queries with the terms suggested by an ad-hoc multi-dimensional topic map. Further, ad-hoc topic maps enable the exploration and analysis of relations between individual topics, which might lead to serendipitous discoveries.
A prototype for monitoring and detecting tax frauds within Danish Social media. The tool provides monitoring of predefined internet sources, search and extraction of relevant content and consequently provides basis for auditing purposes.
I was developing and designing word clouds and recommendation components for the medical personalized event-based surveillance project MECO. The aim of the project was to harvest social media data from different sources and exploit them as complementary reporting mechanism for surveillance experts. The project consortium consisted of several research groups from Europe. In our group in Aalborg, we were focused on adaptive tuning and personalization. (see the final project deliverable). More about projects here.
Enhanced Information Access to Social Streams through Word Clouds with Entity Grouping - Martin Leginus, Leon Derczynski, Peter Dolog WEBIST2015
Beomap: Ad hoc topic maps for enhanced exploration of social media data - Martin Leginus, ChengXiang Zhai, Peter Dolog ICWE2015
Tag cloud generation for results of multiple keywords queries - Martin Leginus, Peter Dolog, Ricardo Lage, ICWE 2013: 233-248 (results)
Graph based techniques for tag cloud generation - Martin Leginus, Peter Dolog, Ricardo Lage, HT 2013: 148-157 (source code + results)
Methodologies for Improved Tag Cloud Generation with Clustering - Martin Leginus, Peter Dolog, Ricardo Lage, Frederico Durao, ICWE 2012: 61-75 (source code + results)
Improving Tensor Based Recommenders with Clustering - Martin Leginus, Peter Dolog, Valdas Zemaitis, UMAP 2012: 151-163 (source code + results)
SimSpectrum: A Similarity Based Spectral Clustering Approach to Generate a Tag Cloud - Frederico Durao, Peter Dolog, Martin Leginus, Ricardo Lage, ICWE Workshops 2011: 145-154
Enhanced Information Access to Social Streams through Word Clouds with Entity Grouping presented at WEBIST in Lisbon.
Predicting Depression via Social Media seminar at Machine intelligence group in Aalborg
Graph based techniques for tag cloud generation presented at ACM HT2013 in Paris, France
Tag cloud generation for results of multiple keywords queries presented at ICWE2013, Aalborg.
Poster and short presentation at ICWE2013 on Tag and word clouds as means of navigation support in social systems.
Relational Clustering for Multi-type Entity Resolution seminar at Machine intelligence group in Aalborg
Methodologies for Improved Tag Cloud Generation with Clustering presented at ICWE2012 in Berlin, Germany
Improving tensor based recommenders with clustering presented at UMAP2012 in Montreal, Canada
Introduction to Matrix and Tensor factorization seminar at IWIS group in Aalborg
Algorithms for Estimating Relative Importance in Networks seminar at Machine intelligence group in Aalborg
Online-Updating Regularized Kernel Matrix Factorization Models for Large-Scale Recommender Systems. seminar at Machine intelligence group in Aalborg