Computer Science EE Topic Ideas + Examples

It is time to start writing your Computer Science extended essay and you can't find a topic that suits you? Don't worry, we've got you! Below you can find a list of 10 different EE topics with linked exemplars to help you start the writing process. Good luck!
 

Computer Science EE Research Question Generator

Try our RQ Generator for free. Enter a topic and we'll instantly generated research questions that perfectly align with IB criteria.

Generator

Describe your initial idea or enter keywords (optional)

Describe your initial idea or enter keywords (optional)

Generate

Machine learning models

Decision-making algorithms

Image classification

Data structures

Data processing

Computer Science EE ideas
 

1. Machine learning models for predicting stock prices: The emergence of machine learning models has popularized the prediction of stock closing prices, driven by technological advancements and the availability of historical data. Retail investors and companies alike utilize these models to gain insights into future stock prices. This research aims to assess the efficiency of machine learning models: Linear Regression, Random Forest, and K-Nearest Neighbor commonly used for this purpose, comparing their performance and identifying contributing factors.
   
    

2. K-means Unsupervised Learning: Clustering, which identifies patterns in unlabeled data by grouping similar objects, has wide-ranging applications from customer behavior analysis to categorizing wine types. This study evaluates the performance of K-means clustering, focusing on configurable parameters like initial placement, number of clusters, and features, using silhouette score and iteration count as metrics. By examining how these parameters influence clustering effectiveness, industries such as businesses stand to benefit from improved data organization and insights. 
   
    

3. Relationship between languages and compression algorithms: This study focuses on how computer science technologies strive to reduce data size, particularly examining the impact of language on compression efficiency. By investigating the effectiveness of Entropy and Dictionary-based algorithms on various languages using Huffman and LZW methods, the research aims to determine how language influences compression. Understanding which languages compress best could lead to the optimization of information storage, highlighting the importance of language selection in data compression. 
   
    

4. Accuracy of backtracking on decision-making in Tic-tac-toe: Programming this game requires data structures such as 2D arrays, trees, stacks, and queues. Decision-making techniques like backtracking, alpha-beta pruning, and breadth-first search are used by the AI to predict and make accurate moves, enhancing its chances of winning. The research aims to explore the extent to which using backtracking in a binary tree improves decision-making accuracy in Tic-Tac-Toe, crucial for understanding the effectiveness of algorithms in improving AI performance. Additionally, evaluating the impact of backtracking can lead to insights into resource utilization and processing efficiency, contributing to advancements in artificial intelligence and decision-making algorithms.
   
    

5. Convolutional Neural Networks and telescopic images: Neural Networks (NNs), such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), simulate the human brain's data processing, leading to advancements in Computer Vision (CV) and Natural Language Processing (NLP). Telescopes collect vast amounts of image data, requiring precise analysis by astronomers, which can be time-consuming and resource-intensive. Crowdsourced initiatives like Galaxy Zoo provide large datasets for morphological classification, making them ideal for Deep-learning technology applications. This study aims to evaluate the efficiency of different models, including CNN architectures like ResNet and DenseNet, and Decision Trees, in processing large telescopic image datasets for morphological classification, addressing a key challenge in astronomy and potentially revolutionizing data processing in the field.
   
    

   
    

6. Artificial Intelligence in sports: Within AI, neural networks are particularly relevant for applications such as judicial AI, which aims to replace human judges in sports. The implementation of AI in sports judging seeks to address issues of bias and bribery, exemplified by past scandals like the 2002 Winter Olympics figure skating scandal, ultimately aiming to enhance fairness and accuracy in sports adjudication. Despite existing technological aids for judges, bias, and inaccuracies persist, underscoring the need for AI-driven solutions to mitigate such issues and improve decision-making objectivity in sports. This EE investigates to what extent may Artificial Intelligence replace human judges in sports.
   
    

7. Use of computer vision to translate American Sign Language: American Sign Language (ASL) is a widely spoken sign language with its own grammar and lexicon, yet its global speakers are limited. In situations where ASL is necessary, such as emergencies, written communication may not suffice, leading to frustration and potential isolation. Computer vision and deep learning offer promising avenues for understanding and translating ASL, leveraging visual inputs to derive meaningful information and facilitate dynamic communication with ASL users. This investigation aims to explore the efficacy of computer vision and deep learning in enabling such communication.
   
    

8. Interpolation and Jump Search Algorithms: Searching, a fundamental operation in computer science, involves locating or determining the location of an element within a collection of data items. Evolving from a basic human activity, search algorithms were developed to efficiently handle massive data sets. These algorithms, such as Interpolation Search and Jump Search, utilize different data structures like arrays and have various applications from databases to artificial intelligence. The research question aims to investigate how the position of the key and different array sizes impact the runtime efficiency of these search algorithms, which is essential in optimizing search operations.
   
    

9. Usage of predictive neural networks in reinforcement learning: While RL has been successful in scenarios with predictable future states, like board games, it faces challenges in domains where outcomes are uncertain, like autonomous driving. To address this, neural networks are proposed to predict future states, enhancing decision-making accuracy. This essay aims to investigate the effectiveness of predictive neural networks in reinforcement learning, particularly in scenarios like autonomous driving, where understanding agent decisions is critical for safety. The research will involve an experiment where an agent learns to balance an inverted pendulum, analyzing training time, predictability, and the impact of prediction accuracy on effectiveness.
   
    

10. Insertion function and structure of Binary Search Trees: This essay delves into the insertion function and structure of Binary Search Trees (BSTs): Scapegoat Tree and Adelson-Velskii and Landis Tree. The investigation centers on comparing the time complexity of inserting values into these two types of BSTs. The analysis is significant as many algorithms, like Huffman Coding, rely on BSTs' rebalancing algorithms to efficiently handle sets of data, especially in scenarios where unbalanced trees can lead to inefficiencies in searching for specific values. The research question addresses the extent of differences in time complexity between the insertion processes of the Scapegoat Tree and AVL Tree.