ORCID Profile
0000-0001-6619-9751
Current Organisations
Gulf of Maine Research Institute
,
University of New South Wales
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-08-2019
DOI: 10.35940/IJEAT.F8848.088619
Abstract: South Korea is a country that has been the center of world's attention because of its music and drama progress. The Korean language is a language that Korean people use to communicate with each other. Hangeul is an alphabet used to write the Korean language. Learning application is a way to introduce the Korean Language to people who want to learn the language therefore, we built an application called Learn Hangeul. It contains an introduction to the alphabet and basic words, and also equipped with a quiz feature. Fisher-Yates Shuffle method is used for shuffling the questions in the application’s Quiz feature. User satisfaction after using the application be measured using a Likert scale, and the result in usefulness, satisfaction, and ease of learning are satisfied.
Publisher: Private Company Technology Center
Date: 28-02-2023
DOI: 10.15587/1729-4061.2023.271822
Abstract: In this study, we develop a web-based Decision-Making Tool (DMT) based on a four-year research project in building the proper multiple criteria Decision-Making Framework (DMF) for infrastructure project selection automation. Several challenges in selecting and prioritizing infrastructure projects include poor front-end planning, lack of project funding, improper investment, unsustainable development, regulatory barriers, and poor coordination among stakeholders. The Non-Structural Fuzzy Decision Support System II (NSFDSS-II) is chosen as the main method applied in the proposed DMF since it could resolve complex multi-criteria problems, even without sufficient information provided. When developing the DMT, Agile software development method is used since the development cycle can be run in a light and fast manner with iterative and incremental processes. The DMT is successfully developed by using PHP, HTML, and JavaScript which implement the proposed NSFDSS-II method. We further tested the decision results from the DMT by using eight real past infrastructure projects from relevant infrastructure agencies in Indonesia, such as the Ministry of Public Works and Housing (MPWH), the Ministry of Transportation, and the Local Government. The DMT outcomes were compared with the actual implementation status and evaluated by an independent expert. It was found that the decision results from the developed DMT are in accordance with the real implementation status of evaluated projects. The DMT is recommended to be used for infrastructure project selection automation. However, despite of its fast and accurate result, the DMT should be tested on larger number of infrastructure projects in the future
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-08-2019
DOI: 10.35940/IJITEE.J9738.0881019
Abstract: The Marketing ision of Universitas Multimedia Nusantara has a role in attracting prospective students to join and continue their next level of education. To help their task, the Marketing ision needs an application that can predict the site traffic profile visits for the next following day. Therefore, a website-based application was created to assist the Marketing ision. We used the Weighted Moving Average algorithm to calculate the number of subsequent visitors and its future prediction number. We also used the Weighted Mean Absolute Percentage Error after getting the prediction results to get the error rate of a prediction made. In the implementation process that has been done, the percentage error is 2.57%. The built application also had been evaluated by the UMN Marketing ision employees by the means of questionnaires to find out how satisfied the application was.
Publisher: The World Academy of Research in Science and Engineering
Date: 15-02-2020
Publisher: IEEE
Date: 2017
Publisher: IEEE
Date: 10-2019
Publisher: IEEE
Date: 11-2017
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-08-2019
DOI: 10.35940/IJEAT.F8142.088619
Abstract: Human Development Index (HDI) is an indicator to see the ruling government achievement, especially in the human development aspect. It was introduced by United Nations Development Programme (UNDP) in their 1990 report and integrated three basic dimensions of human development, i.e., the ability to lead a long and healthy life, to acquire knowledge, and to achieve a decent standard of living. Indonesia itself was ranked 113 out of 188 UNDP countries and ranked fifth among ASEAN countries for HDI value based on UNDP Human Development Re-port 2016: Human Development for Everyone. In this study, we are trying to forecast Indonesia’s HDI using Exponentially Weighted Moving Average (EWMA) method. EWMA is a popular and widely used method to smooth out random fluctuations. Based on the experiments conducted in 34 provinces in Indonesia, EWMA has been successfully applied to forecast the human development index in Indonesia. It has an average score of MSE and MAPE values at 1.002279153 for MSE and 1.374754513% for MAPE.
Publisher: IGI Global
Date: 04-2019
Abstract: Time series are one of the most common data types encountered by data scientists and, in the context of today's exponentially increasing data, learning how to best model them to derive meaningful insights is an important skill in the Big Data and Data Science toolbox. As a result, many researchers have dedicated their efforts to developing time series analysis methods to predict future values based on previously observed values. One of the well-known methods is the Holt-Winters' seasonal method, which is commonly used to capture the seasonality effect in time series data. In this study, the authors aim to build upon the Holt-Winters' additive method by introducing new formulas for finding the initial values. Obtaining more accurate estimations of the initial values could result in a better forecasting result. The authors use the basic principle found in the weighted moving average method to assign more weight to the most recent data and combine it with the original initial conditions found in the Holt-Winters' additive method. Based on the experiment performed, the authors conclude that the new formulas for finding the initial values in the Holt-Winters' additive method could give a better forecasting when compared to the traditional Holt-Winters' additive method and the weighted moving average method in terms of the accuracy level.
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-10-2019
DOI: 10.35940/IJEAT.A1150.109119
Abstract: In university, students’ complaints about the services provided are essential things to note because it could lead to higher student transfer if not appropriately handled. In Universitas Multimedia Nusantara (UMN), students can express their complaints through Dewan Keluarga Besar Mahasiswa (DKBM) UMN. By technology development, an online complaint submission system can be applied at UMN. A method that can be used in supporting efficient complaints processing is by using automated classification system since it can save both time and human resources. UMN e-complaints application with automated complaints classification feature was built using the CodeIgniter framework. The evaluation of the e-complaint application was conducted by using the USE Questionnaire. The results show that both DKBM UMN and students agree that the application is useful, easy to use, easy to learn, and satisfying.
Publisher: The World Academy of Research in Science and Engineering
Date: 15-03-2020
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-09-2019
DOI: 10.35940/IJRTE.C4663.098319
Abstract: The capital market is an organized financial system consisting of commercial banks, intermediary institutions in the financial sector and all outstanding securities. One of the benefits of the capital market is creating opportunities for the community to participate in economic activities, especially in investing. In daily stock trading activities, stock prices tend to have fluctuated. Therefore, stock price prediction is needed to help investors make decisions when they want to buy or sell their shares. One asset for investment is shares. One of the stock price indices that attracts many investors is the LQ45 stock index on the Indonesian stock exchange. One of the algorithms that can be used to predict is the k-Nearest Neighbors (kNN) algorithm. In the previous study, kNN had a higher accuracy than the moving average method of 14.7%. This study uses kNN regression method because it predicts numerical data. The results of the research in making the LQ45 stock index prediction application have been successfully built. The highest accuracy achieved reaches 91.81% by WSKT share.
Publisher: Wiley
Date: 29-11-2021
DOI: 10.1111/FAF.12630
Abstract: In a changing climate, there is an imperative to build coupled social‐ecological systems—including fisheries—that can withstand or adapt to climate stressors. Although resilience theory identifies system attributes that supposedly confer resilience, these attributes have rarely been clearly defined, mechanistically explained, nor tested and applied to inform fisheries governance. Here, we develop and apply a comprehensive resilience framework to examine fishery systems across (a) ecological, (b) socio‐economic and (c) governance dimensions using five resilience domains: assets, flexibility, organization, learning and agency. We distil and define 38 attributes that confer climate resilience from a coupled literature‐ and expert‐driven approach, describe how they apply to fisheries and provide illustrative ex les of resilience attributes in action. Our synthesis highlights that the directionality and mechanism of these attributes depend on the specific context, capacities, and scale of the focal fishery system and associated stressors, and we find evidence of interdependencies among attributes. Overall, however, we find few studies that test resilience attributes in fisheries across all parts of the system, with most ex les focussing on the ecological dimension. As such, meaningful quantification of the attributes’ contributions to resilience remains a challenge. Our synthesis and holistic framework represent a starting point for critical application of resilience concepts to fisheries social‐ecological systems.
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-11-2019
DOI: 10.35940/IJITEE.A4440.119119
Abstract: This paper describes the creation of an expert system that is used to diagnose diseases in chili plants using a web-based certainty factor method. This expert system is made based on the weight of symptoms by experts so that users can find out the disease suffered by chili plants based on symptoms that arise. The symptom weight was calculated using the certainty factor method. This method is used to accommodate uncertainties that are often expressed by experts when detecting a disease. The design and manufacture of this expert system are done using the PHP programming language, MySQL database, CSS and the CodeIgniter framework. The results of the analysis between expert and expert systems have 87.09% compatibility level, and based on the calculation of the feasibility of the system with the Delone and McLean model shows that 77.58% of respondents agree Pakarcabaiku.com is a successful system for detecting chili disease.
Publisher: Institute of Advanced Engineering and Science
Date: 10-2020
Abstract: Indonesia is ranked the top 8th out of the total country population in the world for the global spammers. Web-based spam filter service with the REST API type can be used to detect email spam in the Indonesian language on the email server or various types of email server applications. With REST API, then there will be data exchange between the applications with JSON data type using existing HTTP commands. One type of spam filter commonly used is Bayesian Filtering, where the Naïve Bayes algorithm is used as a classification algorithm. Meanwhile, the N-gram method is used to increase the accuracy of the implementation of the Naïve Bayes algorithm in this study. N-gram and Naïve Bayes algorithms to detect spam email in the Indonesian language have successfully been implemented with accuracy around 0.615 until 0.94, precision at 0.566 until 0.924, recall at 0.96 until 1.00, and F-measure at 0.721 until 0.942. The best solution is found by using the 5-gram method with the highest score of accuracy at 0.94, precision at 0.924, recall at 0.96, and F-measure value at 0.942.
Publisher: IEEE
Date: 08-2017
Publisher: Springer Science and Business Media LLC
Date: 05-04-2023
DOI: 10.1007/S00262-023-03430-6
Abstract: CD73 upregulation in tumors leads to local immunosuppression. This phase I, first-in-human study evaluated oleclumab (MEDI9447), an anti-CD73 human IgG1λ monoclonal antibody, alone or with durvalumab in patients with advanced colorectal cancer (CRC), pancreatic ductal adenocarcinoma (PDAC), or epidermal growth factor receptor-mutant non-small-cell lung cancer (NSCLC). Patients received oleclumab 5–40 mg/kg (dose-escalation) or 40 mg/kg (dose-expansion) intravenously every 2 weeks (Q2W), alone (escalation only) or with durvalumab 10 mg/kg intravenously Q2W. 192 patients were enrolled, 66 during escalation and 126 (42 CRC, 42 PDAC, 42 NSCLC) during expansion. No dose-limiting toxicities occurred during escalation. In the monotherapy and combination therapy escalation cohorts, treatment-related adverse events (TRAEs) occurred in 55 and 54%, respectively, the most common being fatigue (17 and 25%). In the CRC, PDAC, and NSCLC expansion cohorts, 60, 57, and 45% of patients had TRAEs, respectively the most common were fatigue (15%), diarrhea (9%), and rash (7%). Free soluble CD73 and CD73 expression on peripheral T cells and tumor cells showed sustained decreases, accompanied by reduced CD73 enzymatic activity in tumor cells. Objective response rate during escalation was 0%. Response rates in the CRC, PDAC, and NSCLC expansion cohorts were 2.4% (1 complete response [CR]), 4.8% (1 CR, 1 partial response [PR]), and 9.5% (4 PRs), respectively 6-month progression-free survival rates were 5.4, 13.2, and 16.0%. Oleclumab ± durvalumab had a manageable safety profile, with pharmacodynamic activity reflecting oleclumab’s mechanism of action. Evidence of antitumor activity was observed in tumor types that are generally immunotherapy resistant. Clinicaltrials.gov, NCT02503774 date of registration, July 17, 2015.
Publisher: JMIR Publications Inc.
Date: 03-07-2023
DOI: 10.2196/43154
Abstract: Tuberculosis (TB) was the leading infectious cause of mortality globally prior to COVID-19 and chest radiography has an important role in the detection, and subsequent diagnosis, of patients with this disease. The conventional experts reading has substantial within- and between-observer variability, indicating poor reliability of human readers. Substantial efforts have been made in utilizing various artificial intelligence–based algorithms to address the limitations of human reading of chest radiographs for diagnosing TB. This systematic literature review (SLR) aims to assess the performance of machine learning (ML) and deep learning (DL) in the detection of TB using chest radiography (chest x-ray [CXR]). In conducting and reporting the SLR, we followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A total of 309 records were identified from Scopus, PubMed, and IEEE (Institute of Electrical and Electronics Engineers) databases. We independently screened, reviewed, and assessed all available records and included 47 studies that met the inclusion criteria in this SLR. We also performed the risk of bias assessment using Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2) and meta-analysis of 10 included studies that provided confusion matrix results. Various CXR data sets have been used in the included studies, with 2 of the most popular ones being Montgomery County (n=29) and Shenzhen (n=36) data sets. DL (n=34) was more commonly used than ML (n=7) in the included studies. Most studies used human radiologist’s report as the reference standard. Support vector machine (n=5), k-nearest neighbors (n=3), and random forest (n=2) were the most popular ML approaches. Meanwhile, convolutional neural networks were the most commonly used DL techniques, with the 4 most popular applications being ResNet-50 (n=11), VGG-16 (n=8), VGG-19 (n=7), and AlexNet (n=6). Four performance metrics were popularly used, namely, accuracy (n=35), area under the curve (AUC n=34), sensitivity (n=27), and specificity (n=23). In terms of the performance results, ML showed higher accuracy (mean ~93.71%) and sensitivity (mean ~92.55%), while on average DL models achieved better AUC (mean ~92.12%) and specificity (mean ~91.54%). Based on data from 10 studies that provided confusion matrix results, we estimated the pooled sensitivity and specificity of ML and DL methods to be 0.9857 (95% CI 0.9477-1.00) and 0.9805 (95% CI 0.9255-1.00), respectively. From the risk of bias assessment, 17 studies were regarded as having unclear risks for the reference standard aspect and 6 studies were regarded as having unclear risks for the flow and timing aspect. Only 2 included studies had built applications based on the proposed solutions. Findings from this SLR confirm the high potential of both ML and DL for TB detection using CXR. Future studies need to pay a close attention on 2 aspects of risk of bias, namely, the reference standard and the flow and timing aspects. PROSPERO CRD42021277155 www.crd.york.ac.uk rospero/display_record.php?RecordID=277155
Publisher: The World Academy of Research in Science and Engineering
Date: 15-10-2019
Publisher: Universitas Ahmad Dahlan
Date: 09-2017
Publisher: Universitas Ahmad Dahlan
Date: 02-2019
Publisher: IEEE
Date: 06-2018
Publisher: The World Academy of Research in Science and Engineering
Date: 25-04-2020
Publisher: The Institute for Research and Community Services (LPPM) ITB
Date: 03-10-2017
Publisher: Elsevier BV
Date: 10-2019
Publisher: IEEE
Date: 08-2019
Publisher: The World Academy of Research in Science and Engineering
Date: 25-04-2020
Publisher: International Journal of Technology
Date: 31-10-2017
Publisher: Institute of Advanced Engineering and Science
Date: 12-2022
DOI: 10.11591/IJAI.V11.I4.PP1596-1606
Abstract: Presently, the Forex market has become the world’s largest financial market with more than US$5 trillion daily volume. Therefore, it attracts many researchers to learn its traded currency pairs characteristics and predict their future values. Here, we propose simple three layers Bidirectional long short-term memory (Bi-LSTM) networks for Forex forecasting with four different merge modes. Moreover, the proposed model is also compared to the conventional long short-term memory (LSTM) networks with the same architecture. Five major Forex currency pairs, namely AUD/USD, EUR/USD, GBP/USD, USD/CHF, and USD/JPY, with more than ten years of historical records are considered in this study. It is revealed from the experimental results that among four available merge modes, the concatenation mode as the default merge mode in Bi-LSTM networks is actually the less preferred mode for Forex forecasting (Root mean square error 0.30685517, mean absolute error 0.27442235, mean absolute percentage error 0.827108%). Moreover, Bi-LSTM average mode gets the highest R2 score that could achieve 89.579%. Therefore, the proposed three layers Bi-LSTM networks could provide a baseline result for developing a good trading strategy in Forex forecasting.
Publisher: Springer International Publishing
Date: 22-08-2020
Publisher: Institute of Advanced Engineering and Science
Date: 02-2019
DOI: 10.11591/IJECE.V9I1.PP485-495
Abstract: span lang="EN-GB" In driving, the most important thing to be considered is safety. The incident problem can happen anytime and anywhere without anyone knowing it before, especially in Cikarang where the level of incident is quite high. It is very important for the public community to act and respond quickly at the time on and around the occurring incident. Nolong.in application is an exact solution because it can send real time notifications to its users. This response is meant to assist in iduals who have been using Nolong.in to help other users who need urgent help when the time of incident in order to prevent unwanted situations. In this research, push notification method has been implemented by using Firebase Cloud Messaging (FCM). The result of the research is the notification click ratio is 90.91% and the click time is 1 minute and 27 seconds. Based on the questionnaires’ results given to the community conclude that Nolong.in application is very beneficial for Gojek Cikarang community's safety. /span
Publisher: IEEE
Date: 11-2017
Publisher: IEEE
Date: 10-2019
Publisher: IEEE
Date: 10-2019
Publisher: Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Date: 30-10-2019
DOI: 10.35940/IJITEE.L3726.1081219
Abstract: Instant messaging has changed and simplified the way people communicate, whether in professional or personal life. Most communication is done through instant messaging, and it is common for people to miss important information. This is due to the huge amount of incoming message notifications, so users tend to accidentally ignore them. This is also experienced by Universitas Multimedia Nusantara (UMN) student committees who communicate via LINE instant messenger. This research showed LINE bot was made by using the Naive Bayes algorithm to classify between important messages and unimportant messages on the committee group. The Naive Bayes algorithm is a classification algorithm based on probability and statistical methods. The Naive Bayes algorithm is chosen because it is widely implemented in spam filtering the method is simple and has good accuracy. The classification process is done by calculating the probability of chat in each class based on the value of the word likelihood which generated in the training process. This research produces spam precision and spam recall as 94.2% and 95.6% respectively.
Publisher: Springer Science and Business Media LLC
Date: 28-04-2022
DOI: 10.1186/S40537-022-00601-7
Abstract: As a new type of currency introduced in the new millennium, cryptocurrency has established its ecosystems and attracts many people to use and invest in it. However, cryptocurrencies are highly dynamic and volatile, making it challenging to predict their future values. In this research, we use a multivariate prediction approach and three different recurrent neural networks (RNNs), namely the long short-term memory (LSTM), the bidirectional LSTM (Bi-LSTM), and the gated recurrent unit (GRU). We also propose simple three layers deep networks architecture for the regression task in this study. From the experimental results on five major cryptocurrencies, i.e., Bitcoin (BTC), Ethereum (ETH), Cardano (ADA), Tether (USDT), and Binance Coin (BNB), we find that both Bi-LSTM and GRU have similar performance results in terms of accuracy. However, in terms of the execution time, both LSTM and GRU have similar results, where GRU is slightly better and has lower variation results on average.
Publisher: IEEE
Date: 2017
Publisher: Universitas Ahmad Dahlan
Date: 03-2017
Publisher: Elsevier BV
Date: 03-2023
Publisher: JMIR Publications Inc.
Date: 10-2022
Abstract: uberculosis (TB) was the leading infectious cause of mortality globally prior to COVID-19 and chest radiography has an important role in the detection, and subsequent diagnosis, of patients with this disease. The conventional experts reading has substantial within- and between-observer variability, indicating poor reliability of human readers. Substantial efforts have been made in utilizing various artificial intelligence–based algorithms to address the limitations of human reading of chest radiographs for diagnosing TB. his systematic literature review (SLR) aims to assess the performance of machine learning (ML) and deep learning (DL) in the detection of TB using chest radiography (chest x-ray [CXR]). n conducting and reporting the SLR, we followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A total of 309 records were identified from Scopus, PubMed, and IEEE (Institute of Electrical and Electronics Engineers) databases. We independently screened, reviewed, and assessed all available records and included 47 studies that met the inclusion criteria in this SLR. We also performed the risk of bias assessment using Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2) and meta-analysis of 10 included studies that provided confusion matrix results. arious CXR data sets have been used in the included studies, with 2 of the most popular ones being Montgomery County (n=29) and Shenzhen (n=36) data sets. DL (n=34) was more commonly used than ML (n=7) in the included studies. Most studies used human radiologist’s report as the reference standard. Support vector machine (n=5), k-nearest neighbors (n=3), and random forest (n=2) were the most popular ML approaches. Meanwhile, convolutional neural networks were the most commonly used DL techniques, with the 4 most popular applications being ResNet-50 (n=11), VGG-16 (n=8), VGG-19 (n=7), and AlexNet (n=6). Four performance metrics were popularly used, namely, accuracy (n=35), area under the curve (AUC n=34), sensitivity (n=27), and specificity (n=23). In terms of the performance results, ML showed higher accuracy (mean ~93.71%) and sensitivity (mean ~92.55%), while on average DL models achieved better AUC (mean ~92.12%) and specificity (mean ~91.54%). Based on data from 10 studies that provided confusion matrix results, we estimated the pooled sensitivity and specificity of ML and DL methods to be 0.9857 (95% CI 0.9477-1.00) and 0.9805 (95% CI 0.9255-1.00), respectively. From the risk of bias assessment, 17 studies were regarded as having unclear risks for the reference standard aspect and 6 studies were regarded as having unclear risks for the flow and timing aspect. Only 2 included studies had built applications based on the proposed solutions. indings from this SLR confirm the high potential of both ML and DL for TB detection using CXR. Future studies need to pay a close attention on 2 aspects of risk of bias, namely, the reference standard and the flow and timing aspects. ROSPERO CRD42021277155 www.crd.york.ac.uk rospero/display_record.php?RecordID=277155
No related grants have been discovered for Seng Hansun.