Feature Selection for Speech Emotion Recognition in Spanish and Basque: On the Use of Machine Learning to Improve Human-Computer Interaction

Study of emotions in human-computer interaction is a growing research area. This paper shows an attempt to select the most significant features for emotion recognition in spoken Basque and Spanish Languages using different methods for feature selection. RekEmozio database was used as the experimental data set. Several Machine Learning paradigms were used for the emotion classification task. Experiments were executed in three phases, using different sets of features as classification variables in each phase. Moreover, feature subset selection was applied at each phase in order to seek for the most relevant feature subset. The three phases approach was selected to check the validity of the proposed approach. Achieved results show that an instance-based learning algorithm using feature subset selection techniques based on evolutionary algorithms is the best Machine Learning paradigm in automatic emotion recognition, with all different feature sets, obtaining a mean of 80,05% emotion recognition rate in Basque and a 74,82% in Spanish. In order to check the goodness of the proposed process, a greedy searching approach (FSS-Forward) has been applied and a comparison between them is provided. Based on achieved results, a set of most relevant non-speaker dependent features is proposed for both languages and new perspectives are suggested.

Ultrastrong light-matter interaction in quantum technologies

120 p.

Specific Interaction with Cardiolipin Triggers Functional Activation of Dynamin-Related Protein 1

Dynamin-Related Protein 1 (Drp1), a large GTPase of the dynamin superfamily, is required for mitochondrial fission in healthy and apoptotic cells. Drp1 activation is a complex process that involves translocation from the cytosol to the mitochondrial outer membrane (MOM) and assembly into rings/spirals at the MOM, leading to membrane constriction/division. Similar to dynamins, Drp1 contains GTPase (G), bundle signaling element (BSE) and stalk domains. However, instead of the lipid-interacting Pleckstrin Homology (PH) domain present in the dynamins, Drp1 contains the so-called B insert or variable domain that has been suggested to play an important role in Drp1 regulation. Different proteins have been implicated in Drp1 recruitment to the MOM, although how MOM-localized Drp1 acquires its fully functional status remains poorly understood. We found that Drp1 can interact with pure lipid bilayers enriched in the mitochondrion-specific phospholipid cardiolipin (CL). Building on our previous study, we now explore the specificity and functional consequences of this interaction. We show that a four lysine module located within the B insert of Drp1 interacts preferentially with CL over other anionic lipids. This interaction dramatically enhances Drp1 oligomerization and assembly-stimulated GTP hydrolysis. Our results add significantly to a growing body of evidence indicating that CL is an important regulator of many essential mitochondrial functions.

ETCS protokoloaren hedatzea simulazio hibridodun ingurune batean System-in-the-loop erreminta erabiliz.

[EU]Gaur egun, Europa mailan European Rail Traffic Management System (ERTMS) seinaleztapen-sistema bateratua hedatzen ari dira trenbide sare desberdinen arteko elkar eragintasuna bultzatzeko. Proiektu honen helburua da ERTMS sistemaren barneko ETCS protokoloa hedatzea simulazio hibridodun ingurune batean, ERTMS sistemaren hedatzea azkartuko duten erakusleak sortuz. Horretarako, OPNET simulagailuaren System-in-the-loop erreminta erabili da. Erreminta hau baliatuz ETCS protokoloaren pakete errealak ingurune simulatuan integratzeko funtzioen liburutegi bat idatzi da. Amaitzeko, liburutegi hori baliatuz ETCS protokoloak sareko arazoen aurrean duen errendimenduaren analisi bat burutu da eta liburutegi berri horrek pakete errealak simulatuetara itzultzean (eta kontrakoa) duen errendimendua zein den aztertu da.