IJIEEB Vol. 15, No. 2, Apr. 2023
Cover page and Table of Contents: PDF (size: 640KB)
Chatbots are a technological leap in conversational services, generating messages to users either following a set of rules to respond based on recognized patterns or training themselves from previous data or conversations. The primary goal is to enable a device to communicate with a user upon receiving natural language user requests using artificial intelligence and machine learning to generate automated responses. Technology is progressively catering to the questions, both in academic and business contexts, such as situations that require agents to investigate the cause of customer dissatisfaction or to recommend products and services. Significance of this research is to reduce the human dependency and improving customer support by providing close to human natural responses using pattern matching and deep learning on the custom-made data. The main objective of this work is to (a) study the existing literature on cutting-edge technologies in chatbot development in terms of research trends, legacy components, techniques, datasets, and domains specifically in e-commerce and (b) to develop a product that fill some of the gaps/missing functionality identified in current frameworks. We have achieved the following, (a) generated small yet generic dataset, which can be used for all types of products, (b) the intents are identified accurately by the bot using deep learning, whenever a user query.[...] Read more.
Since the last 5 years, digital economy is growing steadily in Indonesia. Right now, the digital economy faces some potential problems and Covid-19 pandemic. This paper presents current data of the national Gross Domestic Product (GDP) and other GDPs (billion IDR) and the number of start-up, and predicts near some categories of future GDP and numbers of available new start-up for the next few years. The forecast will use Markov chain analysis. The results indicate that, while there are problems faced by the digital economy industry, the GDP and numbers of start-up are significantly increasing.[...] Read more.
Malware classification has already been a prominent concern for decades, and malware attacks have proliferated at an astounding rate, constituting a significant threat to cyberspace. Deep learning (DL) and malware image approaches are becoming more prevalent in the field of malware analysis, with spectacular results. This work focuses on the challenge of classifying malware variants that are represented as images. This study employs visualization and proposes a convolutional neural network (CNN) based DL model to effectively and accurately classify malware. The proposed model is trained and tested on a very challenging and heterogeneous dataset, and it achieves accuracy of 98.179%, precision of 97.39%, a F1-score of 97.70%, and a fast classification speed (3 seconds needed to test 934 unseen malware). This demonstrates the proposed model's incredibly quick, effective and accurate performance. The proposed model outperformed existing traditional DL models in terms of various performance measures and demonstrated its usefulness in classifying malware families through visualization. This study and experimental results reveal that small-scale malware images and a simple CNN architecture alone are capable of accurately classifying malware families with high classification accuracy.[...] Read more.
In December 2019, the Novel Coronavirus became a global epidemic. Because of COVID-19, all ongoing plans had been postponed. Lockdowns were imposed in areas where there was an excessive number of patients. Constantly locking down areas had a significant negative influence on the economy, particularly on developing and underdeveloped countries. But the majority of countries were locking down their areas without making any assumptions where some were successful and some were failures. In this situation, this paper presents a novel approach for determining which parts of a country should be immediately placed under lockdown during any pandemic situation while considering the lockdown history at the time of COVID-19. This work makes use of a self-established dataset containing data from several countries of the world and uses the successful presence of lockdown in that area as the target attribute for machine learning algorithms to determine the areas to keep under lockdown in the future. Here, the Random Forest algorithm has provided the highest accuracy of 92.387% indicating that this model can identify the areas with an impressive level of accuracy to retain under lockdown.[...] Read more.
The majority of projects fail to achieve their intended objectives, according to research. This could arise for a number of reasons, such as ensuring requirements are managed, excessive documentation of the code, or the difficulty in delivering software that includes all the requested features on time. An effort could be made to overcome such failure rates by establishing a proper management of requirements and concept of reusability. The correct requirements can be identified by checking similarity between the requirements received from the various stakeholders. A reusable software component can result in substantial savings in both time and money. It can be challenging to make a choice regarding the reuse of certain software components. A comparison of the requirements of a new project with those of previous projects prior to starting a new project or even at a later stage during development is useful for identifying reusable components. This paper proposes a framework (ReSim) for identifying software requirements' similarities, in an attempt to improve reusability and identify the correct requirements. A crucial component of ReSim is to measure similarity between software requirements. Different well-known similarity measurement techniques used by the researchers to evaluate the similarity between the software requirements. Some of the methods used to measure this include dice, jaccard, and cosine coefficients, but in this paper, we have used recently developed hybrid method which considers not only semantic information including lexical databases, word embeddings, and corpus statistics, but also implied word order information and produced significant improvements in the results related to the measurement of semantic similarity between words and sentences. As part of the experiments, the study used PURE dataset - in order to demonstrate the efficacy of the proposed framework. As a result, recently developed hybrid method of measuring the requirements similarity is more accurate than Dice, Jaccard, and Cosine, while Cosine is a better choice than Dice, and Jaccard is more accurate than Dice. Thus, ReSim outperforms existing approaches when tested on the PURE dataset, providing the most accurate results for both functional and non-functional requirements.[...] Read more.