Abstract

Abstract
This paper is about long-term travel time prediction in public transportation. However, it can be useful for a wider area of applications. It follows a heterogeneous ensemble approach with dynamic selection. A vast set of experiments with a pool of 128 tuples of algorithms and parameter sets (a&ps) has been conducted for each of the six studied routes. Three different algorithms, namely, random forest, projection pursuit regression and support vector machines, were used. Then, ensembles of different sizes were obtained after a pruning step. The best approach to combine the outputs is also addressed. Finally, the best ensemble approach for each of the six routes is compared with the best individual a&ps. The results confirm that heterogeneous ensembles are adequate for long-term travel time prediction. Namely, they achieve both higher accuracy and robustness along time than state-of-the-art learners.

Abstract
The main goal of recommender systems is to assist users in finding items of their interest in very large collections. The use of good automatic recommendation promotes customer loyalty and user satisfaction because it helps users to attain their goals. Current methods focus on the immediate value of recommendations and are evaluated as such. This is insufficient for long term goals, either defined by users or by platform managers. This is of interest in recommending learning resources to learn a target concept, and also when a company is organizing a campaign to lead users to buy certain products or moving to a different customer segment. Therefore, we believe that it would be useful to develop recommendation algorithms that promote the goals of users and platform managers (e.g. e-shop manager, e-learning tutor, ministry of culture promotor). Accordingly, we must define appropriate evaluation methodologies and demonstrate the concept on practical cases.

Abstract

Abstract

Abstract
Machine learning processes consist in collecting data, obtaining a model and applying it to a given task. Given a new task, the standard approach is to restart the learning process and obtain a new model. However, previous learning experience can be exploited to assist the new learning process. The two most studied approaches for this are metalearning and transfer learning. Metalearning can be used for selecting the predictive model to use over a determined dataset. Transfer learning allows the reuse of knowledge from previous tasks. Our aim is to use metalearning to support transfer learning and reduce the computational cost without loss in terms of performance, as well as the user effort needed for the algorithm selection. In this paper we propose some methods for mapping the transfer of weights between neural networks to improve the performance of the target network, and describe some experiments performed in order to test our hypothesis.