INESC-ID

About: INESC-ID is a nonprofit organization based out in Lisbon, Portugal. It is known for research contribution in the topics: Field-programmable gate array & Control theory. The organization has 932 authors who have published 2618 publications receiving 37658 citations.

Entropic Profiler – detection of conservation in genomes using information theory

Abstract: In the last decades, with the successive availability of whole genome sequences, many research efforts have been made to mathematically model DNA. Entropic Profiles (EP) were proposed recently as a new measure of continuous entropy of genome sequences. EP represent local information plots related to DNA randomness and are based on information theory and statistical concepts. They express the weighed relative abundance of motifs for each position in genomes. Their study is very relevant because under or over-representation segments are often associated with significant biological meaning. The Entropic Profiler application here presented is a new tool designed to detect and extract under and over-represented DNA segments in genomes by using EP. It allows its computation in a very efficient way by recurring to improved algorithms and data structures, which include modified suffix trees. Available through a web interface http://kdbio.inesc-id.pt/software/ep/ and as downloadable source code, it allows to study positions and to search for motifs inside the whole sequence or within a specified range. DNA sequences can be entered from different sources, including FASTA files, pre-loaded examples or resuming a previously saved work. Besides the EP value plots, p-values and z-scores for each motif are also computed, along with the Chaos Game Representation of the sequence. EP are directly related with the statistical significance of motifs and can be considered as a new method to extract and classify significant regions in genomes and estimate local scales in DNA. The present implementation establishes an efficient and useful tool for whole genome analysis.

Humor–Robot Interaction: A Scoping Review of the Literature and Future Directions

Abstract: Humor is a pervasive feature of everyday social interactions that might be leveraged to improve human–robot interactions (HRI). Our goal is to evaluate how the use of humor can improve HRI and enhance the user’s perception of the robot, as well as to derive implications for future research and development of humorous robots. We conducted a systematic search of 7 digital libraries relevant in the areas of HRI and Psychology for papers that were relevant to our goal. We identified 431 records, published between 2000 and August of 2020, of which 12 matched our eligibility criteria. The included studies reported the results of original empirical research that involved direct or video-mediated interaction of humans and robots. Humor seems to have a positive effect in improving the user’s perception of the robot, as well as the user’s evaluation of the interaction. However, the included studies present a number of limitations in their approaches to robotic humor that need to be surpassed before reaching a final verdict on the value of humor in HRI.

Key Phrase Extraction of Lightly Filtered Broadcast News

Abstract: This paper explores the impact of light filtering on automatic key phrase extraction (AKE) applied to Broadcast News (BN). Key phrases are words and expressions that best characterize the content of a document. Key phrases are often used to index the document or as features in further processing. This makes improvements in AKE accuracy particularly important. We hypothesized that filtering out marginally relevant sentences from a document would improve AKE accuracy. Our experiments confirmed this hypothesis. Elimination of as little as 10% of the document sentences lead to a 2% improvement in AKE precision and recall. AKE is built over MAUI toolkit that follows a supervised learning approach. We trained and tested our AKE method on a gold standard made of 8 BN programs containing 110 manually annotated news stories. The experiments were conducted within a Multimedia Monitoring Solution (MMS) system for TV and radio news/programs, running daily, and monitoring 12 TV and 4 radio channels.

Distributed Software Platform for Automation and Control of General Anaesthesia

Abstract: A parallel computer architecture and a distributed software platform for automation and control of general anesthesia is proposed in this paper. The system is a prototype research platform, intended to help on the development, simulation and test of new control algorithms for general anesthesia. It must be safe when used in real tests and flexible enough to allow the integration of new software modules. The system is composed by two computers, with the specific tasks of anesthesia control and process supervision. The platform makes use of TANGO, a specialized framework for distributed control systems, which provides software mechanisms useful to fulfill the project requirements. The architecture and the set of mechanisms proposed in this paper provide a high degree of flexibility to research on control algorithms, while ensuring the safeness of the whole procedure.

Screening diabetes mellitus 2 based on electronic health records using temporal features

Abstract: The prevalence of type 2 diabetes mellitus is increasing worldwide. Current methods of treating diabetes remain inadequate, and therefore, prevention with screening methods is the most appropriate process to reduce the burden of diabetes and its complications. We propose a new prognostic approach for type 2 diabetes mellitus based on electronic health records without using the current invasive techniques that are related to the disease (e.g. glucose level or glycated hemoglobin (HbA1c)). Our methodology is based on machine learning frameworks with data enrichment using temporal features. As as result our predictive model achieved an area under the receiver operating characteristics curve with a random forest classifier of 84.22 percent when including data information from 2009 to 2011 to predict diabetic patients in 2012, 83.19 percent when including temporal features, and 83.72 percent after applying temporal features and feature selection. We conclude that he pathology prediction is possible and efficient using the patient's progression information over the years and without using the invasive techniques that are currently used for type 2 diabetes mellitus classification.