1,720,960 research outputs found
Multimodal Interaction for users with Autism in a 3D Educational Environment
No full textThe paper presents a new multimodal 3D education environment for children with autism. The new multimodal interaction system considers a combination of visual, voice, and textual modalities. In particular, it allows children with autism to access contents through easy iconic symbols designed to guide them into the innovative environment. For that purpose, it has been very important to consider and identify the classes and attributes necessary to correctly describe different users. In the architecture hierarchy three different user profiles have been considered and structured, following the ICF* model (an extension of the WHO International Classification of Functioning, Disability and Health guidelines), and describing both static and dynamic properties. A specific iconic language has been used to enrich and to present the virtual environment. Simultaneous visual, audio, and cognitive stimuli have been carefully used: they could be potential barriers but also rich opportunities for persons with autism. It has not been only a matter of putting information in a virtual space; it has been necessary to design and develop new languages, metaphors, and codes of interaction, in order to reduce the distance between the user and the system. In this case, communication talks via images, sounds, and gestures have been fundamental. The approach of the project takes into account the user model, the user profiles, the personalization, and the experimentation
Multimodal Interaction Experience for Users with Autism in a 3D Environment
No full textThis paper describes the multimodal 3D game-based learning environments, accessible to users with autism, which we designed, built, and tested. Our innovative interaction system proposes a combination of visual, voice, and textual modalities able to guide the user through the 3D environment and allowing her/him to access contents in a multimodal and personalised way. The design process had to consider the different modalities users could have used to access the contents, and this multimodal user interface had implications on accessibility. By taking care of the multimodality of the realised environment, and of the relational context, children and adults with autism can have the unique opportunity of bypassing some of the difficulties they encounter during their social life, such as visual contact, unwieldy movements, and face-to-face interactions. From this point of view, shared 3D virtual environments with multimodal interaction interfaces represent an important opportunity able to improve the social life of persons with autism. The entire structure of the work has been characterised by five main pillars and innovative aspects: multimodal strategies to explore the environment and to access contents; user model; user profiles; personalisation of the contents; experimentation/validation. The proposed user model (based on the ICF* specification, a customised, extended version of the WHO International Classification of Functioning, Disability and Health guidelines - ICF) allowed an extensible, detailed personalisation, as different attributes described users both from the static and the dynamic points of view. Users with different levels of autism were profiled using our model, and involved in the experiment, permitting us to validate the effectiveness of the global approach. The results of such experiment showed us that indeed users appreciated the system and were able to take advantage of all educational opportunities the system provided
Forensic Examinations: Computational Analysis and Information Extraction
No full textComplex interaction analysis, and information extraction from text and speech are an active research field, based on linguistic theories and NLP techniques. In this paper we present a conceptual model that aims at generating a rich description of forensic examinations, exploiting –in a novel way– forensic, psychological, and linguistic theories. Such description is then translated into profiles and a report. Profiles describe examiners and the person under examination, while the report evaluates the whole examination. Our main goal is to provide a didactical tool for improving forensic examination techniques. The model is based on a multi-layered set of HMMs, which leverage and combine speech (from audio recordings) and textual (from related transcriptions) features, classifying the examination at several granularity levels. Then, a rule-based expert system generates profiles and an evaluation of the examination. We also created a new audio/textual corpus based on real examinations collected from Italian trials. DIKE is the prototype we are currently implementing and that we plan to use for model validation
Extracting emotions and communication styles from vocal signals
No full textMany psychological and social studies highlighted the two distinct channels we use to exchange information among us---an explicit, linguistic channel, and an implicit, paralinguistic channel. The latter contains information about the emotional state of the speaker, providing clues about the implicit meaning of the message. In particular, the paralinguistic channel can improve applications requiring human-machine interactions (for example, Automatic Speech Recognition systems or Conversational Agents), as well as support the analysis of human-human interactions (think, for example, of clinic or forensic applications). In this work we present PrEmA, a tool able to recognize and classify both emotions and communication style of the speaker, relying on prosodic features. In particular, communication-style recognition is, to our knowledge, new, and could be used to infer interesting clues about the state of the interaction. We selected two sets of prosodic features, and trained two classifiers, based on the Linear Discriminant Analysis. The experiments we conducted, with Italian speakers, provided encouraging results (Ac=71% for classification of emotions, Ac=86% for classification of communication styles), showing that the models were able to discriminate among emotions and communication styles, associating phrases with the correct labels
Evaluating forensic examinations in a court of law : the DIKE model
No full textIl lavoro svolto ha visto la realizzazione di DIKE (Description of Interrogations by Knowledge Extraction) un tool progettato per analizzare interrogatori di carattere forense, attraverso l’analisi di informazioni altamente rilevanti. Questo lavoro ha come obiettivo la progettazione e l’implementazione di un modello concettuale originale in grado di descrivere l’andamento nel tempo di un interrogatorio forense.
DIKE propone un modello concettuale multi livello, progettato per analizzare molteplici
caratteristiche di un interrogatorio sfruttando congiuntamente teorie, tecniche e
approcci di tre contesti complementari: psicologico, giuridico e linguistico. Per l’implemnetazione del modello concettuale, si è resa necessaria la creazione un nuovo corpus dati fatto da veri interrogatori del contesto giuridico italiano (registrazioni audio e trascrizioni). La struttura dell’intero corpus prevede una suddivisione in frasi etichettate secondo le caratteristiche del modello concettuale. I file audio e le trascizioni sono state allineate attraverso lo sviluppo di algoritmi complessi necessari per gestire contenuti multimediali con bassa qualità di registrazione. Il modello stocastico probabilistico implementato combina caratteristiche audio e testo al fine di ottenere una classificazione automatica degli interrogatori in funzione della sua architettura. DIKE consente di individuare sequenze particolari di informazioni che hanno caratterizzato l’andamento nel tempo dell’interrogatorio, mettendo in evidenza momenti di crisi e/o discontinuità spontanei o provocati attraverso tencniche di interrogatorio particolari. DIKE è un tool didattico utile per migliorare le tecniche di interrogatorio, che consente di generare un ricco set di dati utile per la descrizione dell’interrogatorio e per la definizione dei profili dinamici di comportamento riguardanti sia l’interrogatorio che e i suoi partecipanti. Riassumendo, i contributi principali di questa tesi sono: la definzione di un nuovo modello concettuale originale formato da 11 dimensioni; un nuovo corpus dati fatto da vere registrazioni provenienti dal contesto giuridico italiano; un set di algoritmi sviluppati per l’allineamento audio-testo di contenuti multimediali registrati con bassa qualità; un’istanza del modello concettuale attraverso un approccio stocastico probabilistico basato su HMMs; un tool didattico per il calcolo, la visualizzazione e l’analisi dei profili dell’interragorio e dei suoi partecipanti.This thesis presents DIKE (Description of Interrogations by Knowledge Extraction) a tool that aims at analyzing examinations in a court of law by extracting relevant information. This work aim at the design and implementation of a conceptual model able to describe relevant information, and useful to understand how forensic examinations develop.
DIKE is based on an original multi dimensional conceptual model, which represents several aspects of examinations according to psychological, juridical, and linguistic theories. For implementing such a model, we created a new audio/textual annotated corpus, using real examination recordings and transcriptions coming from Italian trials, and annotated with sentence- and utterance-level labels. Such a corpus permits to automatically annotate new examinations by means of original multi level, HMMbased model. Audio and transcriptions are automatically aligned by means of an original algorithm, conceived for low quality and noisy audio recordings. The multi-level HMM-based model leverages and combines speech and textual features, classifying the examination according to several dimensions. DIKE permits to highlight dialogue sequences where the speakers experienced crises, as a consequence of deliberately provoked or unwanted stressful events. DIKE, as a didactical tool, permits to improve examination techniques, by generating in a novel way a rich description of examinations, used to define a profile for the examination and a profile for each of those partaking in the examination dialogue. Summing up, the main contributions of this thesis are: the definition of a new multi-dimensional conceptual model representing examinations under different points of view; a new audio/textual, annotated corpus composed by real forensics examinations; an alignment algorithm tailored to noisy environments; a specific multi-level, HMM-based model; and a new didactical tool for calculating, visualizing, and analyzing speaker’s and dialogue’s profiles.DIPARTIMENTO DI ELETTRONICA, INFORMAZIONE E BIOINGEGNERIAComputer Science and Engineering27PERNICI, BARBARABONARINI, ANDRE
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Extracting Emotions and Communication Styles from Prosody
No full textAccording to many psychological and social studies, vocal messages contain two distinct channels—an explicit, linguistic channel, and an implicit, paralinguistic channel. In particular, the latter contains information about the emotional state of the speaker, providing clues about the implicit meaning of the message. Such information can improve applications requiring human-machine interactions (for example, Automatic Speech Recognition systems or Conversational Agents), as well as support the analysis of human-human interactions (for example, clinic or forensic applications). PrEmA, the tool we present in this work, is able to recognize and classify both emotions and communication style of the speaker, relying on prosodic features. In particular, recognition of communication-styles is, to our knowledge, new, and could be used to infer interesting clues about the state of the interaction. PrEmA uses two LDA-based classifiers, which rely on two sets of prosodic features. Experimenting PrEmA with Italian speakers we obtained Ac = 71% for emotions and Ac = 86% for communication styles
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
- …
