Lucian Sasu

Autonomic monitoring approach based on CEP and ML for logistic of sensitive goods

The main objective of this work is to develop a framework for supporting the development of applications for logistic companies that transport perishable goods (food and medicines). Reducing the amount of lost and damaged perishable goods during transportation and storage represents a substantial global challenge, which imply the implementation of cold chain monitoring at all levels of the supply chains. The framework contains several components that enable: (1) the real-time monitoring of goods during transportation; (2) forecast the temperatures of parcels; (3) generate real-time alerts/early warning when the product are not stored according to the acceptance criteria.

Bayesian ARTMAP for regression

Bayesian ARTMAP (BA) is a recently introduced neural architecture which uses a combination of Fuzzy ARTMAP competitive learning and Bayesian learning. Training is generally performed online, in a single-epoch. During training, BA creates input data clusters as Gaussian categories, and also infers the conditional probabilities between input patterns and categories, and between categories and classes. During prediction, BA uses Bayesian posterior probability estimation. So far, BA was used only for classification. The goal of this paper is to analyze the efficiency of BA for regression problems. Our contributions are: (i) we generalize the BA algorithm using the clustering functionality of both ART modules, and name it BA for Regression (BAR); (ii) we prove that BAR is a universal approximator with the best approximation property. In other words, BAR approximates arbitrarily well any continuous function (universal approximation) and, for every given continuous function, there is one in the set of BAR approximators situated at minimum distance (best approximation); (iii) we experimentally compare the online trained BAR with several neural models, on the following standard regression benchmarks: CPU Computer Hardware, Boston Housing, Wisconsin Breast Cancer, and Communities and Crime. Our results show that BAR is an appropriate tool for regression tasks, both for theoretical and practical reasons.

GPU accelerated information retrieval using Bloom filters

Information retrieval is a technique used in search engines, advertisement placement and cognitive databases. With increasing amounts of data and stringent response time requirements, improving the underlying implementation of document retrieval becomes critical. To this end, we consider a Bloom filter, a simple randomized data structure that answers membership queries with no false negative and customizable false positive probability. Mainly, we focus on the speed-up of the algorithm by using a Graphics Processing Units (GPU) based implementation. Starting from a regular CPU implementation of the Bloom filter algorithm, we employ different optimization techniques on the two basic Bloom filter operations: mapping and querying. An important speed-up is achieved for both operations: over 300x for mapping, and over 20x for querying. Furthermore, we show that the number of hash functions used during the mapping operation, the number of files, and the number of query words have a significant effect on the execution time and the speed-up.

Fuzzy ARTMAP with relevance factor

An incremental, nonparametric probability estimation procedure using a variation of the Fuzzy ARTMAP (FAM) neural network is introduced. The resulted network, called Fuzzy ARTMAP with relevance factor (FAMR), uses a relevance factor assigned to each sample pair, proportional to the importance of the respective pair during the learning phase. Experimental results have shown that FAMR favorably compares with FAM and probabilistic FAM, both as a classifier and as a probability estimator.

Bayesian ARTMAP prediction of biological activities for potential HIV-1 protease inhibitors using a small molecular dataset

Several neural architectures were successfully used to predict properties of chemical compounds. Obtaining satisfactory results with neural networks depends on the availability of large data samples. However, most classical Quantitative Structure-Activity Relationship studies have been performed on small datasets. Neural models do generally infer with difficulty from such datasets. In our study, we analyze the performance of the Bayesian ARTMAP for the prediction of biological activities of HIV-1 protease inhibitors, when inferring from a small and structurally diverse dataset of molecules. The Bayesian ARTMAP is a neural model which uses both competitive learning and Bayesian prediction, and has both the universal approximation and best approximation properties. It is the first time when this model is used in a “real-world” function approximation application. We compare the performance of the Bayesian ARTMAP to several other models, each implementing a different learning mechanism. Experiments are performed within Wekas “Experimenter” standard environment. For our small and structurally diverse dataset of chemical compounds, the Bayesian ARTMAP is a good prediction tool, and the most accurate prediction models are the ones which perform local approximation.

Parallel Bayesian ARTMAP and Its OpenCL Implementation

The Bayesian ARTMAP neural network, introduced by Vigdor and Lerner, is an incremental learning algorithm which can efficiently process massive datasets for classification, regression, and probabilistic inference tasks. We introduce the parallelized version of the BA neural network and implement it in OpenCL. Our implementation runs on both multi-core CPUs and GPUs architectures. We test the Parallel Bayesian ARTMAP on several classification and regression benchmarks focusing on speedup and scalability. In some cases, the parallel BA runs by an order of magnitude faster than the sequential implementation. Our implementation has the potential to scale for OpenCL devices with increasing number of compute units.

Fault recovery mechanism for smart city environments

In this paper we present the work done to develop a fault recovery component for the situation when the quality of the data streams from a smart city environment drops. The fault recovery component is able to generate estimated values when the stream generates invalid or missing observations. The component is easy to configure and deploy for a large number of data streams from a smart city environment. As a result of that, the fault recovery component contains an incremental learning model, which is able to train itself during component execution. The paper describes the architecture of the fault recovery component and the result of the evaluation done for three candidate incremental learning models.