Effect of information gain on document classification using k-nearest neighbor

State universities have a library as a facility to support students’ education and science, which contains various books, journals, and final assignments. An intelligent system for classifying documents is needed to ease library visitors in higher education as a form of service to students. The documents that are in the library are generally the result of research. Various complaints related to the imbalance of data texts and categories based on irrelevant document titles and words that have the ambiguity of meaning when searching for documents are the main reasons for the need for a classification system. This research uses k-Nearest Neighbor (k-NN) to categorize documents based on study interests with information gain features selection to handle unbalanced data and cosine similarity to measure the distance between test and training data. Based on the results of tests conducted with 276 training data, the highest results using the information gain selection feature using 80% training data and 20% test data produce an accuracy of 87.5% with a parameter value of k=5. The highest accuracy results of 92.9% are achieved without information gain feature selection, with the proportion of training data of 90% and 10% test data and parameters k=5, 7, and 9. This paper concludes that without information gain feature selection, the system has better accuracy than using the feature selection because every word in the document title is considered to have an essential role in forming the classification.

Spatial dynamics model of land use and land cover changes: A comparison of CA, ANN, and ANN-CA

Land use and land cover (LULC) changes through built-up area expansion always increases linearly with land demand as a consequence of population growth and urbanization. Cirebon City is a center for Ciayumajakuning Region that continues to grow and exceeds its administrative boundaries. This phenomenon has led to peri-urban regions which show urban and rural interactions. This study aims to analyze (1) the dynamics of LULC changes using cellular automata (CA), artificial neural network (ANN), and ANN-CA; (2) the influential factors (drivers); and (3) change probability in the period 2030 and 2045 for Cirebon’s peri-urban. We used logistic regression as quantitative approach to analyze the interaction of drivers and LULC changes. The LULC data derived from Landsat series satellite imagery in 1999-2009 and 2009-2019, validation of dynamic spatial model refers to 100 LULC samples. This research shows that LULC changes are dominated by built-up area expansion which causes plantations and agricultural land to decrease. The drivers have a simultaneous effect on LULC changes with r-square of 0.43, where land slope, distance from existing built-up area, distance from CBD, and accessibility are significant triggers. LULC simulation of CA algorithm is the best model than ANN and ANN-CA based on overall accuracy and overall accuracy (0.96, 0.75, 0.73 and 0.95, 0.66, 0.66 respectively), it reveals urban sprawl through the ribbon and compact development. The average probability of built-up area expansion is 0.18 (2030) and 0.19 (2045). If there is no intervention in spatial planning, this phenomenon will decrease productive agricultural lands in Cirebon's peri-urban.

Fuzzy-AHP MOORA approach for vendor selection applications

Vendor selection is a critical activity in order to support the achievement of company success and competitiveness. Significantly, the company has some specific standards in the selection. Therefore, an evaluation is needed to see which vendors match the company's criteria. The purpose of this study is to evaluate and select the proposed vendor in a web-based decision support system (DSS) by using the fuzzy-AHP MOORA approach. The fuzzy-AHP method is used to determine the importance level of the criteria, while the MOORA method is used for alternative ranking. The results showed that vendor 4 has the highest score than other alternatives with a value of 0.2536. Sensitivity analysis showed that the proposed DSS fuzzy-AHP MOORA concept was already solid and suitable for this problem, with a low rate of change.

Software similarity measurements using UML diagrams: A systematic literature review

Every piece of software uses a model to derive its operational, auxiliary, and functional procedures. Unified Modeling Language (UML) is a standard displaying language for determining, recording, and building a software product. Several algorithms have been used by researchers to measure similarities between UML artifacts. However, there no literature studies have considered measurements of UML diagram similarities. This paper presents the results of a systematic literature review concerning similarity measurements between the UML diagrams of different software products. The study reviews and identifies similarity measurements of UML artifacts, with class diagram, sequence diagram, statechart diagram, and use case diagram being UML diagrams that are widely used as research objects for measuring similarity. Measuring similarity enables resolution of the problem domains of software reuse, similarity measurement, and clone detection. The instruments used to measure similarity are semantic and structural similarity. The findings indicate opportunities for future research regarding calculating other UML diagrams, compiling calculation information for each diagram, adapting semantic and structural similarity calculation methods, determining the best weight for each item in the diagram, testing novel proposed methods, and building or finding good datasets for use as testing material.

Block-hash of blockchain framework against man-in-the-middle attacks

Payload authentication is vulnerable to Man-in-the-middle (MITM) attack. Blockchain technology offers methods such as peer to peer, block hash, and proof-of-work to secure the payload of authentication process. The implementation uses block hash and proof-of-work methods on blockchain technology and testing is using White-box-testing and security tests distributed to system security practitioners who are competent in MITM attacks. The analyisis results before implementing Blockchain technology show that the authentication payload is still in plain text, so the data confidentiality has not minimize passive voice. After implementing Blockchain technology to the system, white-box testing using the Wireshark gives the result that the authentication payload sent has been well encrypted and safe enough. The percentage of security test results gets 95% which shows that securing the system from MITM attacks is relatively high. Although it has succeeded in securing the system from MITM attacks, it still has a vulnerability from other cyber attacks, so implementation of the Blockchain needs security improvisation.

Whats drive someone to share their knowledge? Indonesia virtual community case

A virtual community is one form of Community of Practices that resides on the internet. This community consists of people who share problems or interests in specific topics. A lot of knowledge comes from this kind of community where the members shared their knowledge voluntarily. Knowledge itself is a valuable resource that gives the owner have an advantage. This research tried to observe what are the driver of this behavior. Knowing these variables could help the organization in defining knowledge sharing system functionality. The variables collected using PRISMA Systematic Literature Review (SLR) protocols, and the hypotheses tested using Partial Least Squares Structural Equation Modeling (PLS-SEM). From this research, we observed that self-efficacy and supporting others significantly influence someone to share their knowledge.

Electronic document authenticity verification of diploma and transcript using smart contract on Ethereum blockchain

Ethereum is one of the oldest examples of blockchain technology provides a system that converts centralized storage to distributed and records transactions by way of decentralized and not by a centralized system and can be verified by each node, therefore it is suitable for storing fingerprints from official diploma documents and transcripts that are published. Smart contract is needed for making contract transactions to Ethereum with programming code, so contracts such as diplomas and transcripts uploaded on the Ethereum blockchain can distribute and produce diploma validation and the authenticity of transcripts with transaction hash, consensus, and comply with ERC-721 token standardization. The results showed that a sample of 5 electronic documents in pdf format with a transaction speed of 1 second on each file that were published and secured with Ethereum blockchain technology can be easily verified for authenticity, the system proposed and developed by us takes in consideration invalid and failure cases by giving the necessary feedback to the user.

Simulation of TOPSIS calculation in Discrepancy-Tat Twam Asi evaluation model

This research’s main objective was to provide information related to the simulation of each calculation stage of the TOPSIS method used in the Discrepancy-Tat Twam Asi evaluation model. The TOPSIS method is used to find dominant indicators in the Discrepancy-Tat Twam Asi evaluation model that determines the effectiveness of blended learning in ICT Vocational Schools. This research used a quantitative approach. The questionnaires were used as a data collection tool in this study. Questionnaires were distributed to 20 respondents (teachers and students at several ICT Vocational Schools in Bali, Indonesia) for initial data needs and distributed to fourexperts to obtain data on the TOPSIS calculation effectiveness results. The analysis technique in this research was carried out by comparing the percentage of effectiveness test results with the standards. The results of this research showed that the simulation of TOPSIS method calculation in the Discrepancy-Tat Twam Asi evaluation model had run more effectively, as indicated by score was 93.13%. The simulation results showed the dominant indicator that determines the effectiveness of the blended learning implementation was I-2 (the existence of academic community support).

Development of conceptual framework for cyber fraud investigation

The increase in the number of internet users in Indonesia as much as 175.4 million as recorded in the Datareportal.com report and 4.83 billion globally, impact the increase in the number of cyber fraud cases. Data states that 96% of fraud cases are not resolved due to fraud methods carried out online and make it difficult for legal officers to obtain evidence. Previous fraud investigation research mainly focused on fraud detection, so this research focuses on submitting a framework for investigating cyber fraud cases. The cyber fraud case requires a new framework for investigation because in this fraud case, there is digital evidence that is very prone to be damaged, lost, or modified, which makes this case unsolved. This research aims to develop a framework that is expected to help auditors to uncover cases of cyber fraud so that resolved cyber fraud cases can increase. The method used in making this framework uses Jabareen's conceptual framework development method, which consists of 6 stages, namely, Mapping the selected data source, extensive reading and categorizing of the chosen data, Identifying and naming objects, Deconstructing and categorizing the concept, Integrating concept, Synthesis, resynthesis. And make it all sense. The framework for cyber fraud investigation uses 22 digital forensic frameworks and eight frameworks for fraud audit investigations. The results of developing a framework using the Jabareen method resulted in 8 stages, integrating various concepts selected from digital forensics and fraud audits. Evaluation of framework development was carried out by giving limited questionnaires to practitioners and academics, which produced 89% for the feasibility value and needs of the framework and 67% there is no need for changes to the framework being developed.

E-commerce website service quality and customer loyalty using WebQual 4.0 with importance performances analysis, and structural equation model: An empirical study in Shopee

The rapid growth of business models in the form of online shops, Business to Consumer (B2C) and the Consumer to Consumer (C2C) concept, has an impact in the form of intense competition between online business owners. It is not easy to find a solution to this problem. The owner must provide good or even excellent service by improving the website quality so that customers can achieve satisfaction and loyalty. This research aimed to specify the influence of website quality on customer loyalty with satisfaction as an e-commerce website moderation variable and to investigate the WebQual 4.0 attributes that affect this relationship of the model. This study used 104 respondents of Shopee Consumer and analyzed with the WebQual with Importance Performance Analysis (IPA) method and Structural Equation Model (SEM). The result is that the service quality variable significantly influence e-commerce website customer satisfaction variable, customer satisfaction variable has no significant effect on e-commerce website customer loyalty variables, and website service quality variable significantly influence e-commerce website customer loyalty variable. The attributes of WebQual 4.0 that need to be considered by e-commerce management to impact the level of user satisfaction and customer loyalty are “design by the type of website”, “provide simple-to-understand information”, “provide information following the format”, “feel safe to complete transactions”, “the website makes the positive/absolute experience for me”, and “provide detailed information”. The research implications can be taken into consideration for e-commerce management.