Loading...
3 results
Search Results
Now showing 1 - 3 of 3
- Supporting Argumentation Dialogues in Group Decision Support Systems: An Approach Based on Dynamic ClusteringPublication . Conceição, Luís; Rodrigues, Vasco; Meira, Jorge; Marreiros, Goreti; Novais, PauloGroup decision support systems (GDSSs) have been widely studied over the recent decades. The Web-based group decision support systems appeared to support the group decision-making process by creating the conditions for it to be effective, allowing the management and participation in the process to be carried out from any place and at any time. In GDSS, argumentation is ideal, since it makes it easier to use justifications and explanations in interactions between decision-makers so they can sustain their opinions. Aspect-based sentiment analysis (ABSA) intends to classify opinions at the aspect level and identify the elements of an opinion. Intelligent reports for GDSS provide decision makers with accurate information about each decision-making round. Applying ABSA techniques to group decision making context results in the automatic identification of alternatives and criteria, for instance. This automatic identification is essential to reduce the time decision makers take to step themselves up on group decision support systems and to offer them various insights and knowledge on the discussion they are participating in. In this work, we propose and implement a methodology that uses an unsupervised technique and clustering to group arguments on topics around a specific alternative, for example, or a discussion comparing two alternatives. We experimented with several combinations of word embedding, dimensionality reduction techniques, and different clustering algorithms to achieve the best approach. The best method consisted of applying the KMeans++ clustering technique, using SBERT as a word embedder with UMAP dimensionality reduction. These experiments achieved a silhouette score of 0.63 with eight clusters on the baseball dataset, which wielded good cluster results based on their manual review and word clouds. We obtained a silhouette score of 0.59 with 16 clusters on the car brand dataset, which we used as an approach validation dataset. With the results of this work, intelligent reports for GDSS become even more helpful, since they can dynamically organize the conversations taking place by grouping them on the arguments used.
- Fast anomaly detection with locality-sensitive hashing and hyperparameter autotuningPublication . Meira, Jorge; Eiras-Franco, Carlos; Bolón-Canedo, Verónica; Marreiros, Goreti; Alonso-Betanzos, AmparoThis paper presents LSHAD, an anomaly detection (AD) method based on Locality Sensitive Hashing (LSH), capable of dealing with large-scale datasets. The resulting algorithm is highly parallelizable and its implementation in Apache Spark further increases its ability to handle very large datasets. Moreover, the algorithm incorporates an automatic hyperparameter tuning mechanism so that users do not have to implement costly manual tuning. Our LSHAD method is novel as both hyperparameter automation and distributed properties are not usual in AD techniques. Our results for experiments with LSHAD across a variety of datasets point to state-of-the-art AD performance while handling much larger datasets than state-of-the-art alternatives. In addition, evaluation results for the tradeoff between AD performance and scalability show that our method offers significant advantages over competing methods.
- Anomaly Detection on Natural Language Processing to Improve Predictions on Tourist PreferencesPublication . Meira, Jorge; Carneiro, João; Bolón-Canedo, Verónica; Alonso-Betanzos, Amparo; Novais, Paulo; Marreiros, GoretiArgumentation-based dialogue models have shown to be appropriate for decision contexts in which it is intended to overcome the lack of interaction between decision-makers, either because they are dispersed, they are too many, or they are simply not even known. However, to support decision processes with argumentation-based dialogue models, it is necessary to have knowledge of certain aspects that are specific to each decision-maker, such as preferences, interests, and limitations, among others. Failure to obtain this knowledge could ruin the model’s success. In this work, we sought to facilitate the information acquisition process by studying strategies to automatically predict the tourists’ preferences (ratings) in relation to points of interest based on their reviews. We explored different Machine Learning methods to predict users’ ratings. We used Natural Language Processing strategies to predict whether a review is positive or negative and the rating assigned by users on a scale of 1 to 5. We then applied supervised methods such as Logistic Regression, Random Forest, Decision Trees, K-Nearest Neighbors, and Recurrent Neural Networks to determine whether a tourist likes/dislikes a given point of interest. We also used a distinctive approach in this field through unsupervised techniques for anomaly detection problems. The goal was to improve the supervised model in identifying only those tourists who truly like or dislike a particular point of interest, in which the main objective is not to identify everyone, but fundamentally not to fail those who are identified in those conditions. The experiments carried out showed that the developed models could predict with high accuracy whether a review is positive or negative but have some difficulty in accurately predicting the rating assigned by users. Unsupervised method Local Outlier Factor improved the results, reducing Logistic Regression false positives with an associated cost of increasing false negatives.