Douglas de O. Cardoso

A weightless neural network-based approach for stream data clustering

One of the major data mining tasks is to cluster similar data, because of its usefulness, providing means of summarizing large ammounts of raw data into handy information. Clustering data streams is particularly challenging, because of the constraints imposed when dealing with this kind of input. Here we report our work, in which it was investigated the use of WiSARD discriminators as primary data synthesizing units. An analysis of StreamWiSARD, a new sliding-window stream data clustering system, the benefits and the drawbacks of its use and a comparison to other approaches are all presented.

Financial credit analysis via a clustering weightless neural classifier

Credit analysis is a real-world classification problem where it is quite common to find datasets with a large amount of noisy data. State-of-the-art classifiers that employ error minimisation techniques, on the other hand, require a long time to converge, in order to achieve robustness. This paper explores ClusWiSARD, a clustering customisation of the WiSARD weightless neural network model, applied to two different credit analysis real-world problems. Experimental evidence shows that ClusWiSARD is very competitive with Support Vector Machine (SVM) w.r.t. accuracy, with the advantage of being capable of online learning. ClusWiSARD outperforms SVM in training time, by two orders of magnitude, and is slightly faster in test time.

A bounded neural network for open set recognition

Open set recognition is, more than an interesting research subject, a component of various machine learning applications which is sometimes neglected: it is not unusual the existence of learning systems developed on the top of closed-set assumptions, ignoring the error risk involved in a prediction. This risk is strictly related to the location in feature space where the prediction has to be made, compared to the location of the training data: the more distant the training observations are, less is known, higher is the risk. Proper handling of this risk can be necessary in various situation where classification and its variants are employed. This paper presents an approach to open set recognition based on an elaborate distance-like computation provided by a weightless neural network model. The results obtained in the proposed test scenarios are quite interesting, placing the proposed method among the current best ones.

WCDS: A Two-Phase Weightless Neural System for Data Stream Clustering

Clustering is a powerful and versatile tool for knowledge discovery, able to provide a valuable information for data analysis in various domains. To perform this task based on streaming data is quite challenging: outdated knowledge needs to be disposed while the current knowledge is obtained from fresh data; since data are continuously flowing, strict efficiency constraints have to be met. This paper presents WCDS, an approach to this problem based on the WiSARD artificial neural network model. This model already had useful characteristics as inherent incremental learning capability and patent functioning speed. These were combined with novel features as an adaptive countermeasure to cluster imbalance, a mechanism to discard expired data, and offline clustering based on a pairwise similarity measure for WiSARD discriminators. In an insightful experimental evaluation, the proposed system had an excellent performance according to multiple quality standards. This supports its applicability for the analysis of data streams.

Weightless neural networks for open set recognition

Open set recognition is a classification-like task. It is accomplished not only by the identification of observations which belong to targeted classes (i.e., the classes among those represented in the training sample which should be later recognized) but also by the rejection of inputs from other classes in the problem domain. The need for proper handling of elements of classes beyond those of interest is frequently ignored, even in works found in the literature. This leads to the improper development of learning systems, which may obtain misleading results when evaluated in their test beds, consequently failing to keep the performance level while facing some real challenge. The adaptation of a classifier for open set recognition is not always possible: the probabilistic premises most of them are built upon are not valid in a open-set setting. Still, this paper details how this was realized for WiSARD a weightless artificial neural network model. Such achievement was based on an elaborate distance-like computation this model provides and the definition of rejection thresholds during training. The proposed methodology was tested through a collection of experiments, with distinct backgrounds and goals. The results obtained confirm the usefulness of this tool for open set recognition.

Clustering data streams using a forgetful neural model

To cluster a data stream is a more challenging task than its regular batch version, having stricter performance constraints. In this paper an approach to this problem is presented, based on WiSARD, a memory-based artificial neural network (ANN) model. This model functioning was reviewed and improved, in order to adapt it to this task. The experimental results obtained support the use of this system for the analysis of data streams in an informative way.

An Empirical Study of the Influence of Data Structures on the Performance of VG-RAM Classifiers

This work investigates the effect of different data structures on the performance and accuracy of VG-RAM-based classifiers. This weightless neural model is based on RAM nodes having very large address input, what suggests the use of special data structures in order to deal with space and time computational costs. Four different data structures are explored, including the classical one used in recent VG-RAM related literature, resulting in a novel and accurate yet fast setup.

Weightless neuro-symbolic GPS trajectory classification

Abstract This paper presents a framework for dealing with the problem of GPS trajectory classification in the context of the Rio de Janeiro’s public transit system (with hundreds or more classes). Such framework combines the versatile WiSARD classifier with a set of rules defined a priori, resulting in a neuro-symbolic learning system with very interesting characteristics and cutting-edge performance. We also verified the influence of different binarization methods in order to adapt raw data to WiSARD, which feeds from binary data only. These ideas were tested against a large data set of trajectories of buses from the city of Rio de Janeiro. The results confirm the practical applicability of those, since the accomplished performance was as good as that of other state-of-the-art rival methods in most test scenarios.