Idoia Cearreta

Towards an Ontology for Describing Emotions

The study of emotion in human beings has traditionally been a research interest area in disciplines such as psychology and sociology. The appearance of affective computing paradigm has made it possible to include findings from these disciplines in the development of affective interfaces. Still, there is a lack of applications that take emotion related aspects into account. This situation is mainly due to the great amount of proposed theoretical models and the complexity of human emotions. Besides, the importance that mobile computing area is acquiring has made necessary to bear context related aspects in mind. The proposal presented in this paper is based on a generic ontology for describing emotions and their detection and expression systems taking contextual and multimodal elements into account. The ontology is proposed as a way to develop a formal model that can be easily computerized. Moreover, it is based on a standard, the Web Ontology Language (OWL), which also makes ontologies easily shareable and extensible. Once formalized as an ontology, the knowledge about emotions is used in order to make computers more accessible, personalised and adapted to user needs.

Automatically generating tailored accessible user interfaces for ubiquitous services

Ambient Assisted Living environments provide support to people with disabilities and elderly people, usually at home. This concept can be extended to public spaces, where ubiquitous accessible services allow people with disabilities to access intelligent machines such as information kiosks. One of the key issues in achieving full accessibility is the instantaneous generation of an adapted accessible interface suited to the specific user that requests the service. In this paper we present the method used by the EGOKI interface generator to select the most suitable interaction resources and modalities for each user in the automatic creation of the interface. The validation of the interfaces generated for four different types of users is presented and discussed.

An ontology for description of emotional cues

There is a great variety of theoretical models of emotions and implementation technologies which can be used in the design of affective computers. Consequently, designers and researchers usually made practical choices of models and develop ad-hoc solutions that sometimes lack flexibility. In this paper we introduce a generic approach to modeling emotional cues. The main component of our approach is the ontology of emotional cues. The concepts in the ontology are grouped into three global modules representing three layers of emotions’ detection or production: the emotion module, the emotional cue module, and the media module. The emotion module defines emotions as represented with emotional cues. The emotional cue module describes external emotional representations in terms of media properties. The media module describes basic media properties important for emotional cues. Proposed ontology enables flexible description of emotional cues at different levels of abstraction. This approach could serve as a guide for the flexible design of affective devices independently of the starting model and the final way of implementation.

Validating a multilingual and multimodal affective database

This paper summarizes the process of validating RekEmozio, a multilingual (Spanish and Basque) and multimodal (audio and video) affective database. Fifty-seven participants validated a sample of 2,618 videos of facial expressions and 102 utterances in the database. The results replicated previous findings of no significant differences in recognition rates among emotions. This validation has allowed having the audio and video material in the database classified in terms of the emotional category expressed. This normative data has proven to be useful for both training affective recognizers and synthesizers and carrying out empirical studies on emotions by psychologists.

A comparison using different speech parameters in the automatic emotion recognition using feature subset selection based on evolutionary algorithms

Study of emotions in human-computer interaction is a growing research area. Focusing on automatic emotion recognition, work is being performed in order to achieve good results particularly in speech and facial gesture recognition. This paper presents a study where, using a wide range of speech parameters, improvement in emotion recognition rates is analyzed. Using an emotional multimodal bilingual database for Spanish and Basque, emotion recognition rates in speech have significantly improved for both languages comparing with previous studies. In this particular case, as in previous studies, machine learning techniques based on evolutive algorithms (EDA) have proven to be the best emotion recognition rate optimizers.

Application of feature subset selection based on evolutionary algorithms for automatic emotion recognition in speech

The study of emotions in human-computer interaction is a growing research area. Focusing on automatic emotion recognition, work is being performed in order to achieve good results particularly in speech and facial gesture recognition. In this paper we present a study performed to analyze different machine learning techniques validity in automatic speech emotion recognition area. Using a bilingual affective database, different speech parameters have been calculated for each audio recording. Then, several machine learning techniques have been applied to evaluate their usefulness in speech emotion recognition, including techniques based on evolutive algorithms (EDA) to select speech feature subsets that optimize automatic emotion recognition success rate. Achieved experimental results show a representative increase in the success rate.

Extending In-Home User and Context Models to Provide Ubiquitous Adaptive Support Outside the Home

Ubiquitous Computing has proved to be an excellent way of providing technological support for the daily life of people within its range. Ambient Assisted Living (AAL), which is largely based on Ubiquitous Computing, aims at tutoring and supervising elderly people and users with physical, sensory or cognitive disabilities in the performance of routine household activities. AAL’s main aim is to increase the autonomy of dependent people in their daily life by providing them with supportive instructions for everyday routines and warnings about home safety issues. This concept can be extended to public spaces, where ubiquitous accessible services allow people with disabilities to access location-dependent web services (providing maps, addresses, transport schedules, etc.) and local intelligent machines (such as information kiosks or ATMs). This approach allows existing knowledge about the users, their common activities, and their environment to be used to extend the in-home AAL concept to the support of common routines performed outside the home. This chapter surveys the modelling techniques used inside the home and discusses the methodologies required for their extension for out-of-home use, including interoperation and sharing of models.

Toward adapting interactions by considering user emotions and capabilities

Ambient Intelligence and Affective Computing areas are working on trying to make interfaces more natural from the human side. Taking personal, device and system characteristics into account, adaptive interfaces might help to achieve a greater user satisfaction and it is expected the results can be enhanced when bearing user affectivity in mind. In this paper, an ontology that describes affective interactions has been extended in order to represent information about users sensorial and perceptual capabilities when he/she interacts with systems. Two use cases applying ontology are presented herein.

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.

Applying the affinto ontology to develop a text-based emotional conversation system

With the recent spread of computing systems the need to enhance interactions between users and systems is evident. Conversation systems have a key role to play in achieving this. However, further efforts are needed to enhance conversation systems that use text to interact with users. This paper presents a text conversation system that includes user emotion recognition and generation, with the aim of achieving a more natural communication. The Affinto ontology is used to perform these tasks.