Noticias

Hola,

Esta semana tenemos lectura de TFGs y TFMs, estáis todos invitados. Puede ser muy interesante para los que estáis ahora comenzando el TFG o el TFM.

Antes de la lectura subiremos al canal de YouTube los videos.

 

 

 

 

 

 

 

 

 

 

 

Miércoles

31/01/2018

B-225 

10:00

Development of a Real Time Classification System of Twitter Trends based on Machine Learning Techniques

Daniel Mata-Nieves. (2018). Development of a Real Time Classification System of Twitter Trends based on Machine Learning Techniques. Trabajo de fin de carrera (TFG). 

Miércoles

31/01/2018

B-225 

10:25

Design and Implementation of an embodied cognitive conversational agent for assisting in a web site based on the API.AI conversational platform.

Ángel Silván. (2018). Design and Implementation of an embodied cognitive conversational agent for assisting in a web site based on the API.AI conversational platform.. Trabajo de fin de carrera (TFG).

Miércoles

31/01/2018

B-225 

11:50

Development of a Classifier of Radical Tweets using Machine Learning Algorithms

Mario Hernando. (2018). Development of a Classifier of Radical Tweets using Machine Learning Algorithms. Trabajo de fin de carrera (TFG).

Miércoles

31/01/2018

B-225 

12:15

Design and development of a System for Detecting Cyberbullying in Twitter based on Machine Learning Techniques

Jaime Palos. (2018). Design and development of a System for Detecting Cyberbullying in Twitter based on Machine Learning Techniques. Trabajo de fin de carrera (TFG). 

Miércoles

31/01/2018

B-225 

12:40

Design and Development of a Lyrics Emotion Analysis System for Creative Industries

José María Izquierdo-Mora. (2018). Design and Development of a Lyrics Emotion Analysis System for Creative Industries. Trabajo de fin de carrera (TFG).

Miércoles

31/01/2018

B-225 

13:05

Development of a YouTube videos feelings analiser

José Miguel Valle Salas. (2018). Development of a YouTube videos feelings analiser. Trabajo de fin de carrera (TFG). 

Miércoles

31/01/2018

B-225 

13:30

Design and development of a classification system of political messages in social networks using Natural Language Processing and Machine Learning techniques

Carlos Bermúdez Cruces. (2018). Design and development of a classification system of political messages in social networks using Natural Language Processing and Machine Learning techniques. Trabajo de fin de carrera (TFG). 

Jueves

01/02/2018

B-225 

10:00

Design and Implementation of a Google Action enabled Smart Agent System for Mobile App Review Monitoring based on Sentiment Analysis Techniques

Antonio Fernández. (2017). Design and Implementation of a Google Action enabled Smart Agent System for Mobile App Review Monitoring based on Sentiment Analysis Techniques. Trabajo de fin de carrera (PFC). ETSI Telecomunicación, Universidad Politécnica de Madrid.

Jueves

01/02/2018

B-225 

11:15

Design and Evaluation of a Conversational Agent Prototype based on Recurrent Neural Networks using Tensorflow.

Diego San Cristobal. (2018). Design and Evaluation of a Conversational Agent Prototype based on Recurrent Neural Networks using Tensorflow.. Trabajo de fin de carrera (Master Thesis). 

Jueves

01/02/2018

B-225 

11:40

Development of a Deep Learning Based Sentiment Analysis and Evaluation Service

Ignacio Corcuera-Platas. (2018). Development of a Deep Learning Based Sentiment Analysis and Evaluation Service. Trabajo de fin de carrera (Master Thesis). 

 

El Grupo de Innovación Educativa del GSI "Un Enfoque de Aprendizaje basado en Retos para Técnicas en Análisis de Datos", cuyo objetivo es el desarrollo de competiciones de análisis de datos para las asignaturas de Máster en que participa en docencia. Como resultado del proyecto se generarán varios conjuntos de datos etiquetados para docencia así como un sistema de recogida de datos.

Este lunes 15 de enero  de 2018 se ha celebrado la entrega de premios en el Paraninfo de la Universidad Politécnica de la Cátedra Ingeniero General. D. Antonio Remón y Zarco del Valle.

Felicidades a David García, que ha recibido el segundo premio al mejor Trabajo Fin de Grado por sul TFG titulado "Diseño e implementación de un modelo social de redes terroristas basado en agentes y técnicas de análisis de redes sociales".

  

The recent publication Enhancing deep learning sentiment analysis with ensemble techniques in social applications (Araque O., Corcuera-Platas I., Sanchez-Rada J.F., Iglesias C.A.) has reached the Most Downladed category in the journal Expert Systems with Applications (JCR, Q1, Impact Factor 3.928 in 2016).

 

Abstract

Deep learning techniques for Sentiment Analysis have become very popular. They provide automatic feature extraction and both richer representation capabilities and better performance than traditional feature based techniques (i.e., surface methods). Traditional surface approaches are based on complex manually extracted features, and this extraction process is a fundamental question in feature driven methods. These long-established approaches can yield strong baselines, and their predictive capabilities can be used in conjunction with the arising deep learning methods. In this paper we seek to improve the performance of deep learning techniques integrating them with traditional surface approaches based on manually extracted features. The contributions of this paper are sixfold. First, we develop a deep learning based sentiment classifier using a word embeddings model and a linear machine learning algorithm. This classifier serves as a baseline to compare to subsequent results. Second, we propose two ensemble techniques which aggregate our baseline classifier with other surface classifiers widely used in Sentiment Analysis. Third, we also propose two models for combining both surface and deep features to merge information from several sources. Fourth, we introduce a taxonomy for classifying the different models found in the literature, as well as the ones we propose. Fifth, we conduct several experiments to compare the performance of these models with the deep learning baseline. For this, we use seven public datasets that were extracted from the microblogging and movie reviews domain. Finally, as a result, a statistical study confirms that the performance of these proposed models surpasses that of our original baseline on F1-Score.