Computer Science Masters Papers

Permanent URI for this collectionhdl:10365/32552

Browse

Browsing Computer Science Masters Papers by browse.metadata.program "Software Engineering"

Filter results by typing the first few letters
Now showing 1 - 20 of 51
  • No Thumbnail Available
    Item
    Agent Communication and Negotiation in a Supply Chain
    (North Dakota State University, 2013) Upadhyay, Rajat
    Supply Chain Management is an essential management paradigm for almost every organization. Effective implementation of Supply Chain Management (SCM) can significantly increase an organization’s profit. In this paper, a simple SCM is implemented using multi-agent modeling. A typical SCM consists of the supplier, manufacturer, inventory, seller, and customer. A multi-agent system provides a natural solution for SCM because various entities involved in SCM can be represented as intelligent agents. In the present approach various agents representing various entities of SCM, negotiate with each other in order to achieve their goals. Apart from the entities associated with typical SCM, additional entities have been added to the proposed system. This system consists of six agents representing the SCM entities: Customer Agent, Seller Agent, Coordinator Agent, Inventory Agent, Manufacturer Agent, and Supplier Agent. The system also consists of other agents to initialize the system and the database.
  • No Thumbnail Available
    Item
    Assessment and Recommendation of Requirement Management Tool Based on User Needs
    (North Dakota State University, 2012) Mohpal, Aditi
    Over recent years, requirement management has played a very vital role in materializing IT projects. Whether it is the healthcare sector, finance sector, banking sector, or any other industrial sector, requirement management is crucial for the successful completion of IT projects. Due to a tight time to market and severe competition, requirement management becomes one of the most critical activities for the software-development lifecycle process [1]. Requirements change all the time during the software-development lifecycle. Unmanaged changes lead to project overruns. Most companies are using requirement-management tools to manage requirements. However, many companies end up choosing a requirement-management tool that does not address the project’s needs and problems. In this research, requirement-management tools available on the market have been identified, and a “decision-making application” has been developed for systematic evaluation of requirement-management tools based on user needs so that a project team can identify and choose the correct requirement-management solution.
  • No Thumbnail Available
    Item
    Automated Detection of Acute Leukemia Using K-Means Clustering Algorithm
    (North Dakota State University, 2019) Arya, Minakshi
    Detection of ALL can be done through the analysis of white blood cells (WBCs) called leukocytes. Usually, the analysis of blood cells is performed manually by skilled operators, have numerous drawbacks, such as slow analysis, a non-standard accuracy and skill of the operator. Hence many automated systems are using in order to analyze and classify the blood cells. This paper focuses on an automatic system based on image processing algorithms for the classification of blood cells for detection of Acute Lymphocytic Leukemia (ALL). Experiments were ran using 20 models with PCA and seven models namely Medium KNN, Coarse KNN, Cosine KNN, Cubic KNN, Weighted KNN, Ensemble Boosted trees and Ensemble Bagged trees had 99.9% accuracy. These models are evaluated based on the prediction speed, training time, confusion matrix and ROC. Of all models, the weighted KNN classifier is best when using PCA.
  • No Thumbnail Available
    Item
    Automatic Conceptual Feedback on Software Tests
    (North Dakota State University, 2017) Kumar, Pranay
    Students in introductory programming courses face many challenges. While the educators try their best to assist students with these challenges by adopting different approaches, majority of them end up overlooking the significance of software testing activities. However, these activities are crucial for any developer. This paper proposes and then develops a tool, Testing Tutor, developed by us to help improve student’s testing skills by providing them with conceptual feedback on their tests. We believe that such an approach will help students understand basic testing concepts and understand the reasons for their mistakes instead of offering them the option to simply tweak the source/test code at the erroneous location pointed out by test runner or coverage tool. This paper details the process of designing and building a prototype for such a tool and lessons learnt.
  • No Thumbnail Available
    Item
    Best Practices of ERP Implementation
    (North Dakota State University, 2014) Puri, Rahul
    Today Enterprise Resource Planning (ERP) has become a fundamental necessity for the organization. It serves many functional areas and many industries in an integrated fashion, trying to automate operations from inventory control, supply chain management, sales support, manufacturing production as well as scheduling, financial accounting and cost accounting, customer relationship management and human resources. However, successful implementation and deployment of ERP systems is a challenging task which when not executed successfully will not serve the purpose for which it is being implemented. The objective of the study is to examine and analyze the best practices that have to be adopted while implementing ERP. The scope of this research is to focus the best practices and their subsets for implementing ERP in any organization and to highlight important steps that should be taken before and in the process of ERP implementation.
  • No Thumbnail Available
    Item
    Chatting with Confidence: A Review on the Impact of User Interface, Trust, and User Experience in Chatbots, and a Proposal of a Redesigned Prototype
    (North Dakota State University, 2023) Tamimi, Arin
    As artificial intelligence (AI) becomes more prevalent in our daily lives, trust has become a critical issue in ensuring that AI systems are reliable, ethical, and beneficial to society. This paper explores the role of user experience (UX) in shaping users' trust in chat AI. Chat AI has become increasingly popular as a communication tool, but users often struggle with trusting the technology. The paper examines how different design elements, such as conversational style, interface, and feedback mechanisms, affect users' perception of trust in chat AI by analyzing previous literature written in this area. Research demonstrates that UX plays a critical role in users' trust in chat AI, with factors such as transparency, responsiveness, and empathy contributing to higher levels of trust. Using the results found in the research around this topic, a redesigned prototype of a popular chat AI software called chatGPT was created with the help of Figma.
  • No Thumbnail Available
    Item
    Cognitive Heuristics in Software Engineering: Evaluating Confirmation Bias during Software Testing
    (North Dakota State University, 2017) Hegde, Shishir
    Software engineering is a human centric activity and the thought processes of software engineers have influence on the quality of software products. Cognitive scientists have identified human errors known as cognitive heuristics which could impact quality of a software product. However, there is little empirical evidence to substantiate this assertion in software engineering. In this research we study a specific heuristic and evaluate its impact in software testing. One of the factors that lead to poor quality of testing is to only verify if the system works as expected and ignore negative tests. This can be attributed to a heuristic called as confirmation bias which is defined as the tendency of people to verify their hypothesis rather than refuting them. The experiment design evaluates confirmation bias of software testers and measures their quality of testing. The results indicate that testers with low confirmation bias obtain better overall testing results.
  • No Thumbnail Available
    Item
    A Collaborative Agent-Box System
    (North Dakota State University, 2013) Singh, Kunal Kishore
    In this paper I present a multi-agent system that aims at demonstrating software agent’s collaborative behavior when a number of intelligent agents are clustered together within one environment. This study draws inspiration from a research publication titled as Agent Collaboration and Social Networks, authored by R. Sean Bowman and Henry Hexmoor, which aims at demonstrating agent collaboration by implementing a game of agents, boxes and holes. Similar to the strategies discussed in paper and with a few design modifications of my own I have implemented a simulation that starts with agents, boxes and holes placed in a random order on a square grid. The goal of the agents is to capture a box after having some negotiations with other interested agents, carry and drop the identified box into a hole [9]. The agents have to achieve their goal in a minimum possible time by opting for the shortest available route.
  • No Thumbnail Available
    Item
    Consumer Sentiment Analysis Using Twitter
    (North Dakota State University, 2017) Rashid, Rumana
    Sentiment analysis is the task of finding people’s opinions about specific objects/matters. Ordinary people’s opinions affect the decision-making process. Today, there is a massive explosion of “sentiments” available on social media, e.g. Twitter. Twitter is one of the most popular worldwide social-networking services. Twitter is a widely-used way to get people’s opinion about some topics when you read their posts. In this study, a model was coded in R environment and implemented and tested using a large dataset to estimate people’s opinions concerning specific topics that can be used for implementing better decision in market research. To achieve this goal, a large set of Twitter data along with a reference to a specific business was captured and fed to two models namely, Naïve Bayes and Support Vector Machine (SVM) to classify them. Then, I obtained the result at identifying percentage of positive and negative opinions towards the specific business.
  • No Thumbnail Available
    Item
    Credit Card Fraud Detection Predictive Modeling
    (North Dakota State University, 2020) Sharma, Nishant
    Finance fraud is a growing problem with consequences in the financial industry and data mining has been successfully applied to huge volume of complex financial datasets to automate and analyze credit card frauds in online transactions. Data Mining is challenging process due to two major reasons–first, profiles of normal and fraudulent behaviors change frequently and second, card fraud data sets are highly skewed. This paper investigates and checks the performance of Random Forest Classifier, AdaBoost Classifier, XGBoost Classifier and LightGBM Classifier on highly skewed credit card fraud data. Dataset of credit card transactions is sourced from European cardholders containing 284,786 transactions. These techniques are applied on the raw and preprocessed data. The performance of the techniques is evaluated based on accuracy, sensitivity, specificity, precision. The results indicate about the optimal accuracy for Random Forest, AdaBoost, XGBoost and LightGBM classifiers are 85%, 83%, 97.4%, and 93% respectively.
  • No Thumbnail Available
    Item
    Desktop Interactive Activity Diagram Tutorial
    (North Dakota State University, 2015) Sathiaseelan, AnuEvelyn
    Activity diagram plays a vital role in Business Process Modeling as well as System modeling. In this paper, I have proposed a desktop interactive activity diagram tutorial that has been designed for educational purposes to help someone to construct activity diagram with the following features: (1) There is continuous interaction between the user and the system; (2) It provides ongoing feedback and performance updates to the user after every topic within the chapter; (3) the tutorial will generate a “Certificate of excellence” with the percentage scored if only the user has scored above 90 percentages and the certificate will be send to the user’s given email address; (4) This interactive tutorial is extensible and acts like framework in the sense that, the database is designed in such a way that new tutorials can be added to the existing application without any modification to the front end.
  • No Thumbnail Available
    Item
    Development of Learning Objects For Teaching Software Testing Using a Cyber Learning Environment
    (North Dakota State University, 2016) Singh, Ashish Kumar
    This paper is an attempt to extend the database of learning objects on WReSTT (Web Repository of Software Testing Tools) that students enrolled in programming courses can use to improve their understanding of testing concepts and testing skills. The first learning object describes Microsoft unit test on C# language. Second learning object describes data driven unit tests using different data sources, such as database, xml, and csv. Third learning object describes different assert classes of unit test. Finally, multiple quizzes are presented to assess student learning.
  • No Thumbnail Available
    Item
    Dissertation Calculator Tool for Tracking and Managing Dissertation Progress for Graduate Students
    (North Dakota State University, 2019) Bat-Otgon, Bat-Od
    Dissertation Calculator is a tool for graduate students who are working towards their graduate degree at North Dakota State University to manage and track their dissertation progress. It is also a communication and feedback tool between students and their advisers. It was developed using ASP.NET Core Framework. This project is based on Model-View-Controller and Client-Server model. There are 3 different stakeholders which are student, adviser, and admin so each user has a different view. All the information that is created, modified, and deleted by the users is stored on a database and the host application manages the access to it so it is also incorporates Client-Server model. The client side will be any device with a browser. This tool provides essential help for graduate students who wants to make sure they are on track to complete their dissertation and get useful feedback from their adviser in a timely manner.
  • No Thumbnail Available
    Item
    Document Management System
    (North Dakota State University, 2012) Sharma, Aman
    The debate of using paper documents over computer documents has been a long topic of interest, there has been a lot of research done on the topic. Also, experiments have been done to evaluate the performance of users working with computers v/s working manually with no definitive conclusions (Chris Anderson, 2010) (Askwal, 1985) (Noyes and Graland, 2008). However, one cannot deny the time it takes to store and go through papers manually. Every year, new policies help the user understand the guidelines and operational procedures that they have to follow while being employed by a company. Keeping track of such documents as well as assigning tasks to certain users for revising such are tasks that are lengthy and cumbersome, and are not automated. The software proposed in this paper would help an organization maintain documents, reducing unnecessary tasks and the disk space on the machine.
  • No Thumbnail Available
    Item
    Does Domain Knowledge Increase Creativity During Requirements Development: An Empirical Investigation
    (North Dakota State University, 2014) Sharm, Sonu
    To design and build a system/software, we need to understand the business of the organization, so understanding the business is very important for requirement analysis of such system. This requires an effective and efficient use of domain knowledge. In this paper we discuss the results of the empirical study that was carried out to understand the influence of domain knowledge on the creativity during requirement development phase. And then discuss our results, it’s relevance to software organization and some additional findings from the study.
  • No Thumbnail Available
    Item
    Effectiveness of Human Error Abstraction Assist Tool at Improving the Quality of Software Requirements
    (North Dakota State University, 2020) Deosi, Navneet
    The successful completion of any given software project is dependent on error-free Software Requirement Specification (SRS) documents. SRS documents are prone to human errors and faults because their creation involves a large amount of human interaction. This paper reports an experiment study conducted to gauge the accuracy with which the users identify the root causes of software faults in external and self-developed SRS documents. This study primarily focuses on the effective use of the Human Error Abstraction Assist (HEAA) as a tool to help users abstract human errors. HEAA was chosen because it takes all real-life mishaps into consideration and maps them to the different human error classes – slips, lapses, and mistakes. The data collected from this study reports the students’ performance in terms of accuracy and compares it with the results collected from a past study conducted with industry practitioners that also used the HEAA tool for abstracting errors.
  • No Thumbnail Available
    Item
    Energy Data Analyzer System
    (North Dakota State University, 2012) Pradhan, Basudha
    Analyzing the electrical energy load and pricing data is important to make informed decisions about energy needs and regulations. Data from over a decade are publicly available online for many Independent System Operators (ISOs). To analyze such a vast data collection demands efficiency, automation, and querying capabilities. The Energy Data Analyzer System is a Graphical User Interface (GUI)-based application program specific to the New York ISO (NYISO). The analyzer downloads the data files from NYISO’s website and stores their contents in MySQL database tables for querying. The analyzer can also dynamically create certain custom queries to extract information from the database tables based on a user’s selection. The Energy Data Analyzer System, thus, facilitates the study of electricity markets by empowering users with the ability to query data and to analyze data longitudinally.
  • No Thumbnail Available
    Item
    Enterprise Resource Planning Metamodel Test Pattern
    (North Dakota State University, 2015) Syed, Naveed Ahmed
    Enterprise Resource Planning (ERP) system is a complex distributed software solution that costs millions of dollars to implement. Upgrading an ERP metamodel subsystem introduces critical risks in the functionality of all the ERP applications that are part of an ERP system. It is therefore imperative to verify such critical updates to an ERP system thoroughly before they are released to manufacture. The situation is further exacerbated when the ERP system in context happens to be a legacy one without sufficient automated tests. In order to facilitate effective and efficient implementation of ERP metamodel tests, a verification pattern is presented here. Application of this pattern has shown significant savings in project costs and testing effort, increase in coverage, reliability and reusability of test automation to verify different ERP sub-products (for example UI portal, web services), as well as significant increase in productivity of test engineers.
  • No Thumbnail Available
    Item
    Evaluating Different Mobile Web Navigation Styles
    (North Dakota State University, 2013) Kohli, Jyotjeev Singh
    Web browsing is suffering from the problem of not showing information effectively and efficiently on a small screen. The main aim of this paper is to come up with a solution that will help to minimize this issue. We have compared three different styles and have conducted user study to come to a conclusion and analyze which style is better and solved the current issues. The first style that we have introduced is progressive browsing, it is designed with a perspective of what users want to see; the second style is list view, and is the most popular design; the third style is thumbnail, which incorporated the use of images and links. The results of the user study revealed that progressive browsing was least rated as compared to the other styles. There were many contributing factors that lead to this conclusion and the details are documented in this research paper.
  • No Thumbnail Available
    Item
    Evaluating the Usability of Collaboration Tool for Software Inspection Process
    (North Dakota State University, 2019) Kasireddy, Mounisha
    To develop a good software product, the first and foremost step to master is creating complete and accurate software requirements specification document. Requirements documents are created from through research and are documented for inspection. The inspection team work on finding defects at an early stage, to avoid defects passed on to design team, reduce software development time and improve the product quality. Inspection process, when done in conventional method is labor intensive, cost ambitious and takes overwhelming time. But the benefits of inspection outweigh the difficulties. Software Collaboration tool is one such tool to improve inspection process, help produce complete and accurate SRS document. This paper reviews newly developed Software Collaboration tool and conducts a user acceptance study to evaluate tool’s quality attributes, functionalities, experience, usefulness and collect suggestions for future development. This paper employs survey research to determine whether the tool characters and functionalities meet user needs.