Giovani in un'ora - Ciclo di seminari - Prima parte
- Day - Time: 30 November -0001, h.00:00
- Place: Area della Ricerca CNR di Pisa - Room: C-29
- Alejandro David Moreo Fernandez
- Giulio Rossetti (Dipartimento di Informatica - Università degli Studi di Pisa)
Alejandro Moreo - "Random Indexing Kernels"
Abstract: Documents in text classification are typically represented as sparse vectors according to the bag-of-words (BoW) model. Since in BoW each dimension corresponds to a different term, the resulting vector space is of high dimensionality and term-to-term similarity is not directly observable (term vectors are orthogonal to each other). Latent Semantic Analysis (LSA) helps in deriving reduced representations where the semantic similarity among terms is captured. This information can be integrated into support vector machines via LSA-kernels to improve the classification performance. LSA is however known to be computationally expensive. In this presentation, I will be talking about Random Indexing, a lighter alternative to LSA, and the experiments we have conducted using RI-kernels for text classification.
Giulio Rossetti - "NDlib: Modelling and Analyzing Diffusion Processes over Complex Networks
Abstract: The analysis of the dynamics of and on networks represents a hot topic in the Social Network Analysis playground. To support students, teachers, developers, and researchers we developed a novel framework, named NDlib, an environment designed to describe diffusion simulations. NDlib is designed to be a multi-level ecosystem that can be fruitfully used by different user segments. Upon NDlib, we designed a simulation server that allows remote execution of experiments as well as an online visualization tool that abstracts its programmatic interface and makes available the simulation platform to non-technicians.