No abstract

Two important performance indicators for data mining algorithms are accuracy of classification/ prediction and time taken for training. These indicators are useful for selecting best algorithms for classification/prediction tasks in data mining. Empirical studies on these performance indicators in data mining are few. Therefore, this study was designed to determine how data mining classification algorithm perform with increase in input data sizes. Three data mining classification algorithms-Decision Tree, Multi-Layer Perceptron (MLP) Neural Network and Naïve Bayeswere subjected to varying simulated data sizes. The time taken by the algorithms for trainings and accuracies of their classifications were analyzed for the different data sizes. Results show that Naïve Bayes takes least time to train data but with least accuracy as compared to MLP and Decision Tree algorithms.

This paper proposes optimization models of crude oil distillation column for both limited and unlimited feed stock and market value of known products prices. The feed to the crude distillation column was assumed to be crude oil containing naphtha gas, kerosene, petrol and diesel as the light-light key, light key, heavy key and heavy-heavy key respectively. The models determined maximum concentrations of heavy key in the distillate   ] and feed flow rate ( F )that would give maximum profit with minimum purity sales specification constraints of light key in the distillate and heavy key in the bottom as stated above. The feed loading was limited by the reboiler capacity. However, there is need to simulate the optimization models for an existing crude oil distillation column of a refinery in order to validate the models.

This paper proposes a multi-algorithm strategy for card fraud detection. Various techniques in data mining have been used to develop fraud detection models; it was however observed that existing works produced outputs with false positives that wrongly classified legitimate transactions as fraudulent in some instances; thereby raising false alarms, mismanaged resources and forfeit customers' trust. This work was therefore designed to develop a hybridized model using an existing technique Density-Based Spatial Clustering of Applications with Noise (DBSCAN) combined with a rule base algorithm to reinforce the accuracy of the existing technique. The DBSCAN algorithm combined with Rule base algorithm gave a better card fraud prediction accuracy over the existing DBSCAN algorithm when used alone.Key words: Card fraud detection, density-based spatial clustering of applications with noise (DBSCAN), rule base algorithm, data mining.

A major constraint in hospitals and clinics in the interior villages of low resource countries is the access to stable power supply. Decontamination of reusable metal-based surgical tools is an energy-intensive process, the power required for this procedure may not be accessible in many health centres in low resource countries. Hence, decontamination device with low-energy requirements could immensely benefit village clinical settings. The developed sterilizer utilizes a Fresnel lens in a multi-baffle multi-pass chamber to amplify radiation intensity. An intelligent scheme of air passage was achieved using Fuzzy logic control to ensure control of pressure and temperature regime thermal transport within the chamber. The Fuzzy logic controller program was designed on Matlab's fuzzy logic toolbox and simulated with Simulink to evaluate its accuracy. The performance evaluation of the device showed that at an ambient temperature of 27℃ and a solar radiation intensity of 1362 W/m2, the sterilizer was able to sterilize at a temperature of 169.69℃ which is within the range for efficient dry heat sterilization to take place (140℃-170℃). This work has demonstrated that fuzzy logic controlled dry air sterilizer could achieve temperature ~ 150℃ within the heater chamber which may be suitable for sterilizing used surgical equipment.