KIET Journal of Computing and Information Sciences

Stock Prediction for ARGAAM Companies Dataset

Noman Islam, Salis Khizar Khan, Abdul Rehman, Usman Aftab, Darakhshan Syed — Mon, 07 Aug 2023 00:00:00 -0600

Economic forecasting provides excellent profit opportunities and is a major motivator for most researchers in this field. In the fast-growing business world, the behavior of stock prediction is challenging for most stockholders and commercial investors. It provides benefits to investors to invest more confidently. Machine learning is an emerging technology that provides the capability to learn on its own through real-world intercommunications. Regression is the fundamental technique in machine learning which is useful for real-time applications. This paper experiments with stock price prediction effectively by using three machine learning techniques i.e. linear regression, decision tree, and support vector machine. The techniques were applied to the ARAMCO and Saudi Dairy dataset and the performance is evaluated using various parameters such as R2 value, MAPE, and RMSE. The results substantiated the hypothesis.

Deep Learning based Market Basket Analysis using Association Rules

Hamid Ghous, Mubasher Malik, Iqra Rehman — Mon, 07 Aug 2023 00:00:00 -0600

Market Basket Analysis (MBA) is a data mining technique assisting retailers in determining the customer's buying habits while making new marketing decisions as the buyer's desire frequently changes with expanding needs; therefore, transactional data is getting large every day. There is a demand to implement Deep Learning (DL) methods to manipulate this rapidly growing data. In previous research, many authors conducted MBA applying DL and association rules (AR) on retail datasets. AR identifies the association between items to find in which order the customer place items in the basket. AR is only used in mining frequently purchased items from retail datasets. There is a gap in classifying these rules and predicting the next basket item using DL on the transactional dataset. This work proposes a framework using AR as a feature selection while applying DL methods for classification and prediction. The experiments were conducted on two datasets, InstaCart and real-life data from Bites Bakers, which operates as a growing store with three branches and 2233 products. The AR classified at 80,20 and 70,30 splits using CNNN, Bi- LSTM, and CNN-BiLSTM. The results considering simulation at both splits show that Bi-LSTM performs with high accuracy, around 0.92 on the InstaCart dataset. In contrast, CNN-BiLSTM performs best at an accuracy of around 0.77 on Bites Bakers dataset.

Guidelines for the Development of Automated Test Case Designing and Generation Tool(s)

Saima Rafique, Rizwan Bin Faiz — Mon, 07 Aug 2023 00:00:00 -0600

Test case generation is the most intellectually demanding and labor-intensive activity in software testing. It plays the most important role in the quality of software products and reduces testing costs. Among all these automated test case generation is the most emerging and challenging area in today’s research. Currently, several test case designing/generation tools exist but they are disorganized and unstructured since they are implementing various sets of parameters which is hard to categorize that’s why they are still misaligned with the current requirement of the software industry. There exists a real need to have such a Test Case generation framework that aligns functions of Automated Test Case Generation tools with the current requirements of the software industry. Thus aim of this research is to provide guidelines for the development of automated test case designing/generation tools. To do so we have performed a literature survey which identified six common referred tools and existing proposed frameworks. After that, an international questionnaire-based survey had been conducted in software industries to identify their challenges related to test case generation tools and desired parameters. The results indicated that industries are not satisfied with the current tools and need more sophisticated tools. After analysis, it is concluded that software test case generation is a long process that starts from the identification of parameter sets and ends with the test procedures. It is evident from the results that all assembling test sets activities are highly supported (by all variables) by large and medium scale organizations, highly experienced quality engineers, and organizations that are certified by CMMI (at levels 4 and 5) and ISO(9001 and 90003).

SD-ALB: Software Defined Adaptive load balancing in Data Center Network

Riwan Fazal, Syed Mushhad M. Gilani, Muuhammad Junaid Khalid — Mon, 07 Aug 2023 00:00:00 -0600

Network monitoring has crucial importance in data center networks to analyze the behavior of the underlying network. This analysis is used for working on multiple network parameters and load balancing is one of them. This article proposes an adaptive load-balancing approach to balance the load between the servers while changing its behavior with a change in traffic. Software Defined Networking (SDN) provides the single point of network configuration called SDN controller. This approach is facilitating the easy implementation of adaptive load balancing in Data Center Networks. The proposed approach is an extension to the LBBCLT load balancing approach that uses a dynamic probe generator to probe the servers about the response time and link bandwidth. We incorporate a path selection module in it and the path is selected using the Ant Colony Optimization. The results show that the bandwidth consumption and throughput have been improved and servers are receiving the load according to their capacities.

Impediment in Adaptation of Algorithm Trading: A Case of Frontier Stock Exchange

Uroosa Rahat, Ammar Siddiqui, Khurram Pervez, Muhammad Hasan — Mon, 07 Aug 2023 00:00:00 -0600

The global financial markets have been significantly affected by the rapid change in technology. The study is an attempt to get to know the barriers to not adopting algorithmic trading in conventional stock exchanges. This research aims to plan and analytically proposed a model for explaining the reasons why frontier stock exchange traders and investors are hesitant to adopt algorithmic trading as a tool. The research includes variables; Lack of awareness, Trust, Lack of Government interest, unemployment, and unnecessary investment, which were extracted from previously available literature based on the theory of reason and technology acceptance model (TAM). A sample of 50 traders/investors from Pakistan stock markets was taken by using convenience sampling. Data was collected through a questionnaire and analyzed using correlation and linear regression techniques. The results show trust factor is the biggest hurdle in implementing Algorithm Trading which means countries like Pakistan which are following conventional methods for trading in stock markets have great doubts about the efficiency of Algorithm base trading because of the less human interaction and dependency on machines. Fear of miscalculation and the inexperience of data engineers are also one of the reasons conventional stock exchanges are reluctant to adopt algorithm trading. Similarly, variables like Lack of Government interest, unnecessary investment, and employment have a significant effect on the implementation of algorithm trading. Moreover, lack of awareness is the least significant factor, which shows the traders and investors in the Pakistan Stock Exchange are well aware of algorithm trading but the results cannot be generalized to the population due to a limited sample size of the study.