Amber Williams, Bharat Rawal

This research utilizes the k-Nearest Neighbors (k-NN) algorithm to determine what type of fruit appears in an image. The k-NN algorithm is a machine learning algorithm that classifies a new data point on the majority class of its k-nearest neighbors within the feature space. The k-NN algorithm is beneficial for its simplicity, versatility, and effectiveness in classification and regression tasks. Thus, it is a valuable tool for non-linear and complex data patterns.

To begin, select a small dataset of images with features like fruit color, size, and texture (e.g., orange, blueberry). Once the dataset is obtained, the data must be loaded into Python using the pandas library. To load, implement a simple k-NN classifier for fruit classification with sci-kit-learn. Scikit-learn is a machine-learning library for Python that provides tools and algorithms for data analysis and modeling. Following this, split the dataset into training and testing sets to evaluate the model’s performance. From here, experiment with different values of k in the k-NN algorithm to find the best parameters for classification.

The hypothesis of this experiment is the k-NN algorithm will effectively classify fruits based on features such as color, shape, and texture and demonstrate its ability to create an accurate fruit classification system.

Kelvin Iphy, Holden Heldenbrand, Joseph Overstreet

The goal of our research is to explore characteristic patterns (called MindPrint) to distinguish individuals based on brain signal responses collected by an EEG system. The patterns that are collected are unique to the individual, and therefore would provide them a unique key for user authentication. The process involves a series of tests which record the participant’s brain waves in response to certain set stimuli. These include audio and visual stimuli, and extra tests are done to provide us with the data that is needed to remove artifacts. Once the data is imported into the pipeline, it is filtered and separated into bands based on spectral frequency. It is then transformed into mean, median, and normal channels using transformations on the data. From here, our system processes important features about the data such as the mean, median, mode, standard deviation, kurtosis, and skew. This data gets transformed into a matrix of the features which serves as the training and testing data source for our model. Our neural network will be trained on the participant’s data, and once the training is done, it will be able to predict our intended target, the participant’s identification number.

Steele Russell, Paulo Regis

Introductory programming courses struggle to balance establishing sound programming practices with introduction of software develop technologies.  The introduction of software development technologies adds additional overhead to the task of learning the language.  Putting it simply, the more powerful the tool the greater the time and effort required to learn its use.  The primary software development tool is the Integrated Development Environment.  Unless the choice is made to use command line instructions, which arguably introduces a greater obstacle, some Integrated Development Environment must be employed.  Here the balance must be made between ease of use and level of functionality.  The secondary development tool is debatably the version control system.  When making the decision on whether to employ this technology, the overhead of inclusion must be weighed even more carefully.  The use of IntelliJ and Github integration allows for inclusion of these technologies with minimum additional overhead for students.  The observations made following the inclusion of this Intellij/Github integration in introductory level Computer Science courses will be discussed.

Jonathan Smith, Bharat Rawal, Yenumula Reddy

The aim of the green cloud architecture is to lower data center power usage. The key advantage is green cloud computing architecture is that it ensures real-time performance while lowering the insulation-displacement contact (IDC) energy usage (internet data center). The idea is intended to save both money and the environment. The risk to human life posed by e-waste disposal is also predicted to decrease significantly. The cloud computing and green computing will help enterprises to reduce carbon emissions while also providing a productive work environment. Today, green cloud computing and environmental sustainability are critical. In this presentation we discuss Green computing, its need, and techniques to make Green computing, impacts, approaches, advantages and disadvantages.