Two-Stage-Stochastic-Optimization

Minimizing costs in reservoir storage systems has been a challenging problem over the years. Several methods have been used previously to solve this problem. This paper would be considering the Two-Stage Stochastic Programming technique in solving this problem but the objective function would be a quadratic function. This function is the squared difference between release and the demand. Risk is also considered in solving this problem and the results are benchmarked against that of the Fletcher-Ponnambalam method. The results are also compared to that of the deterministic solution and the case of perfect information. The Two-Stage Stochastic Programming applied to a quadratic objective function gave promising results. As the number of scenarios increased, the optimal value approached that of the FP method. The stochastic solution from the TSP was shown to give 16.8% improvement in costs compared to the deterministic solution and the results of EVPI showed that getting perfect information for our problem would be highly beneficial leading to an 86.7% reduction in cost over time.

Time-Series-Modelling-of-Fresh-Food-Prices-Using-ARIMA-models

The price of food products is largely assumed to increase over time. In this study, an analysis is done for fresh food products to see how the prices behave and then an attempt is made to forecast future prices. The case study here is apples. In particular, we consider apple prices received by growers in the United States from 1980 till date. The data spans from January 1980 to August 2018 as this was the only period where there was consistent data for this particular product. The data is obtained from the United States Department of Agriculture Economic research Service. (USDA-ERS). The time series being analyzed here is a monthly time series. An exploratory data analysis is done to understand the characteristics of the data after which a confirmatory data analysis is done to confirm the initial inferences. ARMA models are fit to both the monthly time series, a forecasting experiment is done to check the model accuracy and minimum mean square error forecasts are used to predict future prices.

Extensive-Comparison-of-Machine-Learning-Algorithms-forCardiotocography-Signal-Classification

Cardiotocography (CTG) has been a widely used process to record fetal heart rate (FHR) and uterine contractions (UC) during pregnancy. The results from the CTG is analyzed and used to classify the fetus into one of several morphological patterns or fetal states. This classification has traditionally been done by obstetricians based on standard and approved guidelines but that does not eliminate the tedious nature of the task nor the high probability of classification errors. Recently, machine learning techniques have been used to make these classifications with high accuracy but no extensive comparisons to determine the best model has been done. We carry out predictions for both fetal state and morphological patterns using 7 different models and an ensemble of the best models. We also explore the correlation between the two sets of labels to see how knowledge of one of them could affect the prediction of the other. We then show that our models performed better than those of other researchers who used the UCI data set, the ensemble worked better than the individual models and the correlation between the labels (fetal state and morphological pattern) improved the accuracy predicting one label when the other one is known.

Breast-cancer-Invasive-Ductal-Carcinoma-IDC-detection-from-images

Breast cancer is the most common form of cancer in women, and invasive ductal carcinoma (IDC) is the most common form of breast cancer. Accurately identifying and categorizing breast cancer subtypes is an important clinical task, and automated methods can be used to save time and reduce error.

Hockey-Project

The aim of this project was to create a means to predict the number of points a player would get given other statistics in the data.