Noticias

Hola,

La semana que viene tenemos varias lecturas de trabajo fin de grado, estáis todos invitados:

Lunes 26 de junio

• 10:00 Guillermo Fernández Escobar, Design and Implementation of an Agent-based Crowd Simulation Model for Evacuation of University Buildings Using Python

• 10:20 Guillermo García Menéndez, Design and Implementation of a mobile application based on Cordova framework and web technologies for a University intranet

• 10:40 Diego Benito Sánchez, Design and Development of a Personality Traits Classifier based on Machine Learning Techniques

• 11:00 Enrique Sánchez Tolbaños, Design and Development of a Cognitive Computing em- powered Robot Assistant for Semantic Task Automation in Smart Spaces

• 11:20 Carlos Moro García, Development of a Task Automation Platform based on Semantic Multievent-driven rules

• 11:40 Eduardo Varas Gandara, Design and Development of a Process Mining System for the Optimization of a Loan Approval Process

• 12:00 Álvaro Gericke Parga, Design and Implementation of a Social Marketplace for a Startup based on a Content Management System. Application of Sentiment Analysis for sensing customer feedback

Martes 27 de junio

• 10:00 Enrique Merino Machuca , Design and implementation of an Agent-based Social Simulation Model of Energy Related Occupant Behaviour in Buildings

• 10:20 David García Martín, Design and Implementation of an Agent-based Social Model of Terrorist Networks based on Social Network Analysis Techniques

• 10:40 Pablo Aramburu García, Design and Development of a Sentiment Analysis System on Facebook from Political Domain

• 11:00 Pablo Aznar Delgado, Design and Development of an Agent-based Social Simula- tion Visualization Tool for Indoor Crowd Analytics based on the library Three.js

• 11:20 Carlos Alonso Aguilar, Design and development of a Personality Prediction System based on Mobile-Phone based Metrics

Hola,

Este lunes a las 17:00 en el Salón de grados, tendrá lugar la lectura y defensa de la tesis de Ganggao, estáis todos invitados.

La tesis se titula "Semantic Similarity Analysis and Application in Knowledge Graphs".

Lugar: Salón de grados (Edificio A)

Fecha: 19 de junio de 2017, 17:00

Resumen

The advanced information extraction techniques and increasing availability of linked data have given birth to the notion of large scale Knowledge Graph (KG). With the increasing popularity of KGs containing millions of concepts and entities, the research of fundamental tools studying semantic features of KGs is critical for the development of KG-based applications, apart from the study of KG population techniques. With such focus, this thesis exploits semantic similarity in KGs taking into consideration of concept taxonomy, concept distribution, entity descriptions and categories. Semantic similarity captures the closeness of meanings. Through studying the semantic network of concepts and entities with meaningful relations in KGs, we proposed a novel WPath semantic similarity metric and new graph-based Information Content (IC) computation method. With the WPath and graph-based IC, semantic similarity of concepts can be computed directly and only based on the structural and statistical knowledge contained in KG. The word similarity experiments have shown that the improvement of the proposed methods is statistical significant comparing to conventional methods. Moreover, observing that concepts are usually collocated with textual descriptions, we propose a novel embedding approach to train concept and word embedding jointly. The shared vector space of concepts and words, has provided convenient similarity computation between concepts and words through vector similarity. Furthermore, the applications of knowledge-based, corpus-based and embedding-based similarity methods are shown and compared in the task of semantic disambiguation and classification, in order to demonstrate the capability and suitability of different similarity methods in specific application. Finally, semantic entity search is used as an illustrative showcase to demonstrate higher level of the application consisting of text matching, disambiguation and query expansion. To implement the complete demonstration of entity-centric information querying, we also propose a rule-based approach for constructing and executing SPARQL queries automatically. In summary, the thesis exploits various similarity methods and illustrates their corresponding applications for KGs. The proposed similarity methods and presented similarity based applications would help in facilitating the research and development of applications in KGs.

Los días 6 y 7 se realiza en París la primera revisión del proyecto europeo EmoSpaces, en la oficinas de Thales.

GSI participa representando al consorcio español y presentando algunos de los resultados, como la simulación social en espacios inteligentes mostrada en el vídeo.

Members of GSI has published the article Sematch: Semantic Similarity Framework for Knowledge Graphs, Ganggao Zhu & Carlos A. Iglesias. (2017) . Knowledge-Based Systems, JCR 3.325 (Q1)Sematch is an integrated framework for the development, evaluation and application of semantic similarity in Knowledge Graphs (KGs). The framework provides a number of similarity datasets and tools, and allow users to interact with KGs, and it is available at https://github.com/gsi-upm/sematch. 

El proyecto MixedEmotions (H2020 644632) llega a su fin tras dos años de trabajo entre siete socios internacionales, entre los que se encuentra el GSI. El resultado es una plataforma big data open source de análisis de emociones en diferentes modalidades (p.e. texto, audio y vídeo) basada en microservicios. Entre estas modalidades destaca el uso innovador de contexto social para el análisis de emociones, desarrollado por el GSI.

La arquitectura de la plataforma se compone de más de dieciséis módulos individuales. La integración de los módulos es posible en parte gracias al esquema y la interfaz de basados en NIF y JSON-LD desarrollados inicialmente por el GSI en el framework de servicios de análisis de sentimientos senpy.

Además de la plataforma, el proyecto ha dejado más de 40 publicaciones en conferencias y revistas científicas, dos capítulos en libros, y varias tesis y trabajos de fin de titulación.