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.

Using Empathy to Improve Human-Robot Relationships

Abstract: For robots to become our personal companions in the future, they need to know how to socially interact with us. One defining characteristic of human social behaviour is empathy. In this paper, we present a robot that acts as a social companion expressing different kinds of empathic behaviours through its facial expressions and utterances. The robot comments the moves of two subjects playing a chess game against each other, being empathic to one of them and neutral towards the other. The results of a pilot study suggest that users to whom the robot was empathic perceived the robot more as a friend.

A highly scalable algorithm for the extraction of cis-regulatory regions

Abstract: In this paper we propose a new algorithm for identifying cis-regulatory modules in genomic sequences. In particular, the algorithm extracts structured motifs, defined as a collection of highly conserved regions with pre-specified sizes and spacings between them. This type of motifs is extremely relevant in the research of gene regulatory mechanisms since it can e! ectively represent promoter models. The proposed algorithm uses a new data structure, called box-link, to store the information about conserved regions that occur in a well-ordered and regularly spaced manner in the dataset sequences. The complexity analysis shows a time and space gain over previous algorithms that is exponential on the spacings between binding sites. Experimental results show that the algorithm is much faster than existing ones, sometimes by more than two orders of magnitude. The application of the method to biological datasets shows its ability to extract relevant consensi.

A Data Mining Approach for the Detection of High-Risk Breast Cancer Groups

Abstract: It is widely agreed that complex diseases are typically caused by the joint effects of multiple instead of a single genetic variation. These genetic variations may show very little effect individually but strong effect if they occur jointly, a phenomenon known as epistasis or multilocus interaction. In this work, we explore the applicability of decision trees to this problem. A case-control study was performed, composed of 164 controls and 94 cases with 32 SNPs available from the BRCA1, BRCA2 and TP53 genes. There was also information about tobacco and alcohol consumption. We used a Decision Tree to find a group with high-susceptibility of suffering from breast cancer. Our goal was to find one or more leaves with a high percentage of cases and small percentage of controls. To statistically validate the association found, permutation tests were used. We found a high-risk breast cancer group composed of 13 cases and only 1 control, with a Fisher Exact Test value of 9.7×10− 6. After running 10000 permutation tests we obtained a p-value of 0.017. These results show that it is possible to find statistically significant associations with breast cancer by deriving a decision tree and selecting the best leaf.

A “Deeper” Look at Detecting Cyberbullying in Social Networks

Abstract: As cyberbullying becomes more and more frequent in social networks, automatically detecting it and pro-actively acting upon it becomes of the utmost importance. In this work, a detailed look at the current state-of-the-art in cyberbullying detection reveals that deep learning techniques have seldom been used to tackle this problem, despite growing reputation in other text-based classification tasks. Motivated by neural networks' documented success, three architectures are implemented from similar works: a simple CNN, a hybrid CNN-LSTM and a mixed CNN-LSTM-DNN. In addition, three text representations are trained from three different sources, via the word2vec model: Google-News, Twitter and Formspring. The experiment shows that these models with one of the above embeddings beat other benchmark classifiers (Support Vector Machines and Logistic Regression) both in an unbalanced and balanced version of the same dataset.

Using rituals to express cultural differences in synthetic characters

Abstract: There is currently an ongoing demand for richer Intelligent Virtual Environments (IVEs) populated with social intelligent agents. As a result, many agent architectures are taking into account a plenitude of social factors to drive their agents' behaviour. However, cultural aspects have been largely neglected so far, even though they are a crucial aspect of human societies. This is largely due to the fact that culture is a very complex term that has no consensual definition among scholars. However, there are studies that point out some common and relevant components that distinguish cultures such as rituals and values. In this article, we focused on the use of rituals in synthetic characters to generate cultural specific behaviour. To this end, we defined the concept of ritual and integrated it into an existing agent architecture for synthetic characters. A ritual is seen as a symbolic social activity that is carried out in a predetermined fashion. This concept is modelled in the architecture as a special type of goal with a pre-defined plan. Using the architecture described, and in order to assess if it is possible to express different cultural behaviour in synthetic characters, we created two groups of agents that only differed in their rituals. An experiment was then conducted using these two scenarios in order to evaluate if users could identify different cultural behaviour in the two groups of characters. The results show that users do indeed identify the differences in the two cultures and most importantly that they ascribe the differences to cultural factors.