Browsing by Author "Bernardete Ribeiro"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- DOTS: Drift Oriented Tool SystemPublication . Antunes, Mário; Costa, Joana; Silva, Catarina; Bernardete RibeiroDrift is a given in most machine learning applications. The idea that models must accommodate for changes, and thus be dynamic, is ubiquitous. Current challenges include temporal data streams, drift and non-stationary scenarios, often with text data, whether in social networks or in business systems. There are multiple drift patterns types: concepts that appear and disappear suddenly, recurrently, or even gradually or incrementally. Researchers strive to propose and test algorithms and techniques to deal with drift in text classification, but it is difficult to find adequate benchmarks in such dynamic environments. In this paper we present DOTS, Drift Oriented Tool System, a framework that allows for the definition and generation of text-based datasets where drift characteristics can be thoroughly defined, implemented and tested. The usefulness of DOTS is presented using a Twitter stream case study. DOTS is used to define datasets and test the effectiveness of using different document representation in a Twitter scenario. Results show the potential of DOTS in machine learning research.
- The impact of longstanding messages in micro-blogging classificationPublication . Costa, Joana; Silva, Catarina; Antunes, Mário; Bernardete RibeiroSocial networks are making part of the daily routine of millions of users. Twitter is among Facebook and Instagram one of the most used, and can be seen as a relevant source of information as users share not only daily status, but rapidly propagate news and events that occur worldwide. Considering the dynamic nature of social networks, and their potential in information spread, it is imperative to find learning strategies able to learn in these environments and cope with their dynamic nature. Time plays an important role by easily out-dating information, being crucial to understand how informative can past events be to current learning models and for how long it is relevant to store previously seen information, to avoid the computation burden associated with the amount of data produced. In this paper we study the impact of longstanding messages in micro-blogging classification by using different training timewindow sizes in the learning process. Since there are few studies dealing with drift in Twitter and thus little is known about the types of drift that may occur, we simulate different types of drift in an artificial dataset to evaluate and validate our strategy. Results shed light on the relevance of previously seen examples according to different types of drift.
- Improving Visualization, Scalability and Performance of Multiclass Problems with SVM Manifold LearningPublication . Silva, Catarina; Bernardete RibeiroWe propose a learning framework to address multiclass challenges, namely visualization, scalability and performance. We focus on supervised problems by presenting an approach that uses prior information about training labels, manifold learning and support vector machines (SVMs). We employ manifold learning as a feature reduction step, nonlinearly embedding data in a low dimensional space using Isomap (Isometric Mapping), enhancing geometric characteristics and preserving the geodesic distance within the manifold. Structured SVMs are used in a multiclass setting with benefits for final multiclass classification in this reduced space. Results on a text classification toy example and on ISOLET, an isolated letter speech recognition problem, demonstrate the remarkable visualization capabilities of the method for multiclass problems in the severely reduced space, whilst improving SVMs baseline performance.