Flávio Martins

Multimodal medical information retrieval with unsupervised rank fusion

Modern medical information retrieval systems are paramount to manage the insurmountable quantities of clinical data. These systems empower health care experts in the diagnosis of patients and play an important role in the clinical decision process. However, the ever-growing heterogeneous information generated in medical environments poses several challenges for retrieval systems. We propose a medical information retrieval system with support for multimodal medical case-based retrieval. The system supports medical information discovery by providing multimodal search, through a novel data fusion algorithm, and term suggestions from a medical thesaurus. Our search system compared favorably to other systems in 2013 ImageCLEFMedical.

Using SKOS vocabularies for improving web search

Knowledge organization systems such as thesauri or taxonomies are increasingly being expressed using the Simple Knowledge Organization System (SKOS) and published as structured data on the Web. Search engines can exploit these vocabularies and improve search by expanding terms at query or document indexing time. We propose a SKOS-based term expansion and scoring technique that leverages labels and semantic relationships of SKOS concept definitions. We also implemented this technique for Apache Lucene and Solr. Experiments with the Medical Subject Headings vocabulary and an early evaluation with Library of Congress Subject Headings indicated gains in precision when using SKOS-based expansion compared to pseudo relevance feedback and no expansion. Our findings are important for publishers and consumer of Web vocabularies who want to use them for improving search over Web documents.

Barbara Made the News: Mining the Behavior of Crowds for Time-Aware Learning to Rank

In Twitter, and other microblogging services, the generation of new content by the crowd is often biased towards immediacy: what is happening now. Prompted by the propagation of commentary and information through multiple mediums, users on the Web interact with and produce new posts about newsworthy topics and give rise to trending topics. This paper proposes to leverage on the behavioral dynamics of users to estimate the most relevant time periods for a topic. Our hypothesis stems from the fact that when a real-world event occurs it usually has peak times on the Web: a higher volume of tweets, new visits and edits to related Wikipedia articles, and news published about the event. In this paper, we propose a novel time-aware ranking model that leverages on multiple sources of crowd signals. Our approach builds on two major novelties. First, a unifying approach that given query q, mines and represents temporal evidence from multiple sources of crowd signals. This allows us to predict the temporal relevance of documents for query q. Second, a principled retrieval model that integrates temporal signals in a learning to rank framework, to rank results according to the predicted temporal relevance. Evaluation on the TREC 2013 and 2014 Microblog track datasets demonstrates that the proposed model achieves a relative improvement of 13.2% over lexical retrieval models and 6.2% over a learning to rank baseline.

Inverse square rank fusion for multimodal search

Rank fusion is the task of combining multiple ranked document lists (ranks) into a single ranked list. It is a late fusion approach designed to improve the rankings produced by individual systems. Rank fusion techniques have been applied throughout multiple domains: e.g. combining results from multiple retrieval functions, or multimodal search where several feature spaces are common. In this paper, we present the Inverse Square Rank fusion method family, a set of novel fully unsupervised rank fusion methods based on quadratic decay and on logarithmic document frequency normalization. Our experiments created with standard Information Retrieval datasets (image and text fusion) and image datasets (image features fusion), show that ISR outperforms existing rank fusion algorithms. Thus, the proposed technique has comparable or better performance than existing state-of-the-art approaches, while maintaining a low computational complexity and avoiding the need for document scores or training data.

Improving Cold-Start Recommendations with Social-Media Trends and Reputations.

In recommender systems, the cold-start problem is a common challenge. When a new item has no ratings, it becomes difficult to relate it to other items or users. In this paper, we address the cold-start problem and propose to leverage on social-media trends and reputations to improve the recommendation of new items. The proposed framework models the long-term reputation of actors and directors, to better characterize new movies. In addition, movies popularity are deduced from social-media trends that are related to the corresponding new movie. A principled method is then applied to infer cold-start recommendations from these social-media signals. Experiments on a realistic time-frame, covering several movie-awards events between January 2014 and March 2014, showed significant improvements over ratings-only and metadata-only based recommendations.

A Vertical PRF Architecture for Microblog Search

In microblog retrieval, query expansion can be essential to obtain good search results due to the short size of queries and posts. Since information in microblogs is highly dynamic, an up-to-date index coupled with pseudo-relevance feedback (PRF) with an external corpus has a higher chance of retrieving more relevant documents and improving ranking. In this paper, we focus on the research question:how can we reduce the query expansion computational cost while maintaining the same retrieval precision as standard PRF? Therefore, we propose to accelerate the query expansion step of pseudo-relevance feedback. The hypothesis is that using an expansion corpus organized into verticals for expanding the query, will lead to a more efficient query expansion process and improved retrieval effectiveness. Thus, the proposed query expansion method uses a distributed search architecture and resource selection algorithms to provide an efficient query expansion process. Experiments on the TREC Microblog datasets show that the proposed approach can match or outperform standard PRF in MAP and NDCG@30, with a computational cost that is three orders of magnitude lower.

Analysis of Subtopic Discovery Algorithms for Real-time Information Summarization

The rise of large data streams introduces new challenges regarding the delivery of relevant content towards an information need. This need can be seen as a broad topic of information. By identifying sub-streams within a broader data stream, we can retrieve relevant content that matches the multiple facets of the topic; thus summarizing information, and matching the initial need. In this paper, we propose to study the generation of sub-streams over time and compare various aggregation methods to summarize information. Our experiments were made using the standard TREC Real-Time Summarization (RTS) 2017 dataset.

Jitter Search: A News-Based Real-Time Twitter Search Interface

In this demo we show how we can enhance real-time microblog search by monitoring news sources on Twitter. We improve retrieval through query expansion using pseudo-relevance feedback. However, instead of doing feedback on the original corpus we use a separate Twitter news index. This allows the system to find additional terms associated with the original query to find more “interesting” posts.