Search results for "machine learning."

An Online Metric Learning Approach through Margin Maximization

2011

This work introduces a method based on learning similarity measures between pairs of objects in any representation space that allows to develop convenient recognition algorithms. The problem is formulated through margin maximization over distance values so that it can discriminate between similar (intra-class) and dissimilar (inter-class) elements without enforcing positive definiteness of the metric matrix as in most competing approaches. A passive-aggressive approach has been adopted to carry out the corresponding optimization procedure. The proposed approach has been empirically compared to state of the art metric learning on several publicly available databases showing its potential bot…

Predicting Next Dialogue Action in Emotionally Loaded Conversation

2021

This paper reports on creating a neural network model for prediction of the next action in a dialogue considering conversation history, i.e. entities, context variables and emotion indicators marking emotionally loaded user utterances. Several experiments were performed to see how the information about emotions affects the accuracy of the model. For the purposes of these experiments, a dataset containing 206 dialogs in Latvian in the transport inquiry domain was created containing both neutral and emotionally loaded utterances. To see if the proposed next dialogue action prediction model architecture is suitable for other languages, the original Latvian utterances were translated into Engli…

Autonomous maritime ecosystem : digital concepts and business case : results from the JYU TJTSM54 course on advanced topics on systems development

2019

RANKING DECISION MAKING UNITS BY MEANS OF SOFT COMPUTING DEA MODELS

2011

This paper presents a method for ranking a set of decision making units according to their level of efficiency and which takes into account uncertainty in the data. Efficiency is analysed using fuzzy DEA techniques and the ranking is based on the statistical analysis of cases that include representative situations. The method enables the removal of the sometimes unrealistic hypothesis of a perfect trade-off between increased inputs and outputs. This model is compared with other DEA models that work with imprecise or fuzzy data. As an illustration, we apply our ranking method to the evaluation of a group of Spanish seaports, as well as teams playing in the Spanish football league. We compar…

Soft Computing Methods for Personnel Selection Based on the Valuation of Competences

2014

Personnel selection based on candidates' competences is a difficult task due to the imprecise description of the applicants' competences and to the existence of several experts simultaneously evaluating those attributes. In this context, fuzzy sets theory provides suitable tools for the attainment of the maximum possible information from imprecise data. In this work, personnel selection methods are proposed that rely on the definition of an ideal candidate. Aggregated fuzzy valuations of each candidate are obtained taking into account the individual valuations provided by the experts. Then, candidates are ranked based on their similarity with the ideal candidate. Three different scenarios a…

Sound and reusable components for abstract interpretation

2019

Abstract interpretation is a methodology for defining sound static analysis. Yet, building sound static analyses for modern programming languages is difficult, because these static analyses need to combine sophisticated abstractions for values, environments, stores, etc. However, static analyses often tightly couple these abstractions in the implementation, which not only complicates the implementation, but also makes it hard to decide which parts of the analyses can be proven sound independently from each other. Furthermore, this coupling makes it hard to combine soundness lemmas for parts of the analysis to a soundness proof of the complete analysis. To solve this problem, we propose to c…

Gesture Recognition for Improved User Experience in a Smart Environment

2013

Ambient Intelligence (AmI) is a new paradigm that specifically aims at exploiting sensory and context information in order to adapt the environment to the user's preferences; one of its key features is the attempt to consider common devices as an integral part of the system in order to support users in carrying out their everyday life activities without affecting their normal behavior. Our proposal consists in the definition of a gesture recognition module allowing users to interact as naturally as possible with the actuators available in a smart office, by controlling their operation mode and by querying them about their current state. To this end, readings obtained from a state-of-the-art…

Statistical analysis of latency outcomes in behavioral experiments

2011

In experimental designs of animal models, memory is often assessed by the time for a performance measure to occur (latency). Depending on the cognitive test, this may be the time it takes an animal to escape to a hidden platform (water maze), an escape tunnel (Barnes maze) or to enter a dark component (passive avoidance test). Latency outcomes are usually statistically analyzed using ANOVAs. Besides strong distributional assumptions, ANOVA cannot properly deal with animals not showing the performance measure within the trial time, potentially causing biased and misleading results. We propose an alternative approach for statistical analyses of latency outcomes. These analyses have less distr…

Using privacy-transformed speech in the automatic speech recognition acoustic model training

2020

Automatic Speech Recognition (ASR) requires huge amounts of real user speech data to reach state-of-the-art performance. However, speech data conveys sensitive speaker attributes like identity that can be inferred and exploited for malicious purposes. Therefore, there is an interest in the collection of anonymized speech data that is processed by some voice conversion method. In this paper, we evaluate one of the voice conversion methods on Latvian speech data and also investigate if privacy-transformed data can be used to improve ASR acoustic models. Results show the effectiveness of voice conversion against state-of-the-art speaker verification models on Latvian speech and the effectivene…

Quantum Machine Learning: A tutorial

2021

This tutorial provides an overview of Quantum Machine Learning (QML), a relatively novel discipline that brings together concepts from Machine Learning (ML), Quantum Computing (QC) and Quantum Information (QI). The great development experienced by QC, partly due to the involvement of giant technological companies as well as the popularity and success of ML have been responsible of making QML one of the main streams for researchers working on fuzzy borders between Physics, Mathematics and Computer Science. A possible, although arguably coarse, classification of QML methods may be based on those approaches that make use of ML in a quantum experimentation environment and those others that take…