2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)最新文献

{"title":"Automated and interpretable m-health discrimination of vocal cord pathology enabled by machine learning","authors":"Nabeel Seedat, V. Aharonson, Y. Hamzany","doi":"10.1109/CSDE50874.2020.9411529","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411529","url":null,"abstract":"Clinical methods that assess voice pathologies are typically based on laryngeal endoscopy or audio-perceptual assessment. Both methods have limited accessibility in low-resourced healthcare settings. M-health systems can provide a quantitative assessment and improve early detection in a patient centered care. Automated methods for voice pathologies assessment apply machine learning methods to acoustic, frequency and noise features extracted from sustained phonation recordings and aim to discriminate pathological voices from controls. The machine learning methods in this study are applied to a discriminating between two prevalent vocal pathologies: vocal cord polyp and vocal cord paralysis. The data was acquired by a low-cost recording device in an experiment at a tertiary medical center and the pathologies were clinically labeled. Acoustic and spectral features were extracted and multiple classifiers compared using batched cross validation. The best classifiers were tree-based classifiers, with the Extra Trees classifier providing the best performance with an accuracy of 0.9565 and F1-score of 0.9130. Explainable AI (XAI) and feature interpretability analysis was carried out to allow clinicians to use the features marked as important to clinical care and planning. The most important features were octave-based spectral contrast and MFCCs 0 to 3. The results indicate a feasibility of machine learning to accurately discriminate between different types of vocal cord pathologies.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123071098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reducing Social Media Users’ Biases to Predict the Outcome of Australian Federal Election 2019","authors":"B. Das, M. Anwar, Iqbal H. Sarker","doi":"10.1109/CSDE50874.2020.9411633","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411633","url":null,"abstract":"Several online social networking sites (OSNs) are being used as a medium of expressing any ideas, opinions, thoughts towards any regular or special issues and also for support or even oppose any social or political matters at the same time. At the age of this modern technology, in any country, people would like to post their views regarding any political party during the election period on such emerging OSNs in order to demonstrate their stands upon them. In this paper, we incorporated the tweets relevant to the Australian federal Election 2019, with a view to serve our primary purpose of predicting the outcome of it. We aggregated two efficacious techniques to extract the information from a large Twitter dataset to count a virtual support for each corresponding political group and propose an approach of reducing users’ biases in OSNs to predict outcome of the election more efficiently. Our investigation finds close relevance with the original results of the election published by the Australian Electoral Commission.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122900995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning Approach for Face Recognition from 3D Models Generated by Multiple 2D Angular Images","authors":"Sabrina Jahan Prova, Shegufta Mehzabin, M. Mahmud, Md. Ashraful Alam","doi":"10.1109/CSDE50874.2020.9411541","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411541","url":null,"abstract":"We propose a machine learning approach for face recognition from 3D models generated by multiple 2D angular images that recognizes faces from multiple angle of a 3D face model. The proposed system uses SFM algorithm with SIFT detector, Approximate Nearest Neighbors (ANN) algorithm and RANSAC algorithm to reconstruct 3D from multiple RGB images. Again, it includes AdaBoost Learning algorithm that is used to train model to recognize faces and we used Local Binary Pattern Histogram (LBPH) which marks the pixels of a picture. The proposed system successfully recognizes faces with a deviation angle up to 120°, (i.e., 60° left and 60° right). Additionally, it gives an accuracy of 80% to 100% depending on angular deviation of up to from 0° to 60°. Nevertheless, the rate of accuracy of our proposed system is reversely proportional to the Angular Deviation.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124415733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lepidoptera Classification through Deep Learning","authors":"Xiaotian Jia, Xueting Tan, Guoen Jin, R. Sinnott","doi":"10.1109/CSDE50874.2020.9411382","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411382","url":null,"abstract":"Deep learning for image recognition has received a lot of attention in recent years. In this paper we present a case study using two state-of-the-art deep learning libraries for image classification based on single phase (Single Shot Detection - SSD) and two-phase (Faster Region-based Convolutional Neural Network – Faster-RCNN) deep learning technologies. The case study is based on classification of lepidoptera: an order of species that includes butterflies and moths. We describe the data that was collected that underpinned this work. We also present the results and discuss the challenges with the work. Finally, we outline the implementation of a mobile application used as the client interface to the final solution.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116863215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive review on the influence of social media marketing in harnessing international students to Australia","authors":"Dona Lankika Shamalee Paranawithana, E. Gide, Robert Wu, G. Chaudhry","doi":"10.1109/CSDE50874.2020.9411536","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411536","url":null,"abstract":"Higher Education Institutions (HEI) across the globe are using social media platforms (such as Facebook, YouTube, LinkedIn, Twitter, Instagram) to attract international students in an increasingly competitive market. International students are the third largest source of revenue for the Australian economy. However, a growing awareness of the positive outcomes from good social media relationships between real and potential students and HEI, measuring the effectiveness of social media in this rapidly changing landscape, has not been utilized effectively in academic research in the Australian context. The efforts of academic institutions are uncoordinated across the Australian landscape, and there is a competitive rather than a cooperative approach in Australian marketing. This research paper explores the gap in academic treatment of best practices for online social media marketing in Australia. It is found that research into lived experiences, emotions and behaviours and the meanings individuals attach to them have been systematically undermined in academic literature on this subject. In this study, by using secondary research literature will closely examine on student attitudes which provides a platform to identify the best practices in the HEI. Further the study brings what competitive behaviours by the HEI can support in creation of a digital framework that can benefit the Australian education institutions to attract and retain international students as the most desired destination for higher education.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132195765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Face-off - Classical and Heuristic-based Path Planning Approaches","authors":"Vishal Chand, Avinesh Prasad, K. Chaudhary, B. Sharma, Samlesh Chand","doi":"10.1109/CSDE50874.2020.9411557","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411557","url":null,"abstract":"Robot path planning is a computational problem to find a valid sequence of configurations to move a robot from an initial to a final destination. Several classical and heuristic-based methods exist that can be used to solve the problem. This paper compares the performance of a classical method based on potential field, Lyapunov-based Control Scheme, with those of the standard and stepping ahead Firefly Algorithms. The performance comparison is based on the optimal path distance and time. The results show that the stepping ahead Firefly algorithm finds a shorter path in lesser duration when compared with the Lyapunov-based method. The LbCS also inherently faces the local minima problem when the start, target, and obstacle’s center coordinates are collinear. This problem is solved using the firefly algorithm where the diversification of the fireflies helps escape local minima.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"249 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123033900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Incremental Improvement for Sub-optimal Euclidean TSP Paths Generated by Traditional Heuristics","authors":"A. Barczak, Erik T. Barczak, N. Reyes","doi":"10.1109/CSDE50874.2020.9411580","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411580","url":null,"abstract":"This paper proposes and tests three simple algorithms that are capable of rapidly improving non-optimal paths for the Euclidean Travelling Salesman Problem (TSP).The ETSP is a special case of the more general TSP, but it still plays a very important part of planning and scheduling. For example, it can be used to optimise CNC tool paths. There is no known polynomial solution for the ETSP (nor for any of the other TSP problems), so various approximation heuristics were proposed over the years.Many advancements were made in the past 10 years. For example, CONCORDE [1] and LKH [2] are two of the most important publicly available implementations that are able to find optimal solutions for graphs up to a few thousand nodes. However, the runtime to find the optimal solution is very long for large graphs. In the case where the application can afford an approximate solution, it is convenient to improve one of the classical heuristic solutions, namely, Nearest neighbour, Greedy and Insertion.The cheapest of the improvement heuristics is 2opt, followed by a generalisation of the former, k-opt.The objective of the three proposed algorithms is to find an improved non-optimal suitable solution in a relatively short time, at least much shorter than the well-known 2-opt or k-opt improvement algorithms. This is of course a compromise, as the improvements are much further away from the optimal path. In many cases k-opt approaches can find optimal solutions, but the runtime can be very long.The three proposed algorithms improve an existing heuristic solution in polynomial time. The first algorithm unravels crossing edges in an sub-optimal path. The second algorithm moves nodes that are closer to edges, improving the path. The third algorithm uses a K-nodes optimiser to stretch small parts of the path.The experiments were carried out using the TSPLIB instances [3] for the instances that are Euclidean (and therefore, symmetric). The results show that one can quickly improve paths obtained using one of the well-known polynomial time heuristics. The proposed algorithms yield better paths than 2-opt paths in more than half of the Euclidean TSPLIB instances. Also, the runtime of the proposed algorithms is much shorter than 2-opt for graphs with more than 100 nodes. It is 2 to 10 times faster than running 2-opt.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123664012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sizing and Mission Analysis of Multirole Short-Haul All-Electric Seaplane for Sustainable Aviation","authors":"Nouman Uddin, D. Singh, R. Pant","doi":"10.1109/CSDE50874.2020.9411375","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411375","url":null,"abstract":"Seaplanes are the aircraft that can take-off and land from water. There are lots of emissions of CO2 and GHGs by aviation industry which can be lower down by the use of electric engines. This paper summarizes the sizing and mission analysis of multirole short-haul all-electric seaplane. There are plenty of opportunities for this type of aircraft as the infrastructural requirements for the operation are much lower compared to those for the conventional aircraft. In this study, an electric seaplane is sized to meet twin operational requirements of a multi-role mission, viz., short-haul air transportation (SHAT) and Air Sea Rescue (ASR). It is seen that though the battery energy density required for the critical mission, viz. ASR, is extremely high ($sim$ 2 kWhr/kg) but for SHAT, it is just 0.625 kWhr/kg, for comparable mass of fossil fuel. In comparison to a conventional aircraft using fossil fuels, the all-electric aircraft results in mitigation of 3600 kg CO2 emission in each ASR mission, thus establishing it as a potential solution for sustainable aviation.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124858282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stock Market Prediction using Supervised Machine Learning Techniques: An Overview","authors":"Zaharaddeen Karami Lawal, Hayati Yassin, R. Zakari","doi":"10.1109/CSDE50874.2020.9411609","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411609","url":null,"abstract":"Stock price prediction is one of the most extensively studied and challenging glitches, which is acting so many academicians and industries experts from many fields comprising of economics, and business, arithmetic, and computational science. Predicting the stock market is not a simple task, mainly as a magnitude of the close to random-walk behavior of a stock time series. Millions of people across the globe are investing in stock market daily. A good stock price prediction model will help investors, management and decision makers in making correct and effective decisions. In this paper, we review studies on supervised machine learning models in stock market predictions. The study discussed how supervised machine learning techniques are applied to improve accuracy of stock market predictions. Support Vector Machine (SVM) was found to be the most frequently used technique for stock price prediction due to its good performance and accuracy. Other techniques like Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), Naïve Bayes, Random Forest, Linear Regression and Support Vector Regression (SVR) also showed a promising prediction result.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123948925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cyberbullying Detection on Social Networks Using Machine Learning Approaches","authors":"Md. Manowarul Islam, Md. Ashraf Uddin, Linta Islam, Arnisha Akter, Selina Sharmin, U. Acharjee","doi":"10.1109/CSDE50874.2020.9411601","DOIUrl":"https://doi.org/10.1109/CSDE50874.2020.9411601","url":null,"abstract":"The use of social media has grown exponentially over time with the growth of the Internet and has become the most influential networking platform in the 21st century. However, the enhancement of social connectivity often creates negative impacts on society that contribute to a couple of bad phenomena such as online abuse, harassment cyberbullying, cybercrime and online trolling. Cyberbullying frequently leads to serious mental and physical distress, particularly for women and children, and even sometimes force them to attempt suicide. Online harassment attracts attention due to its strong negative social impact. Many incidents have recently occurred worldwide due to online harassment, such as sharing private chats, rumours, and sexual remarks. Therefore, the identification of bullying text or message on social media has gained a growing amount of attention among researchers. The purpose of this research is to design and develop an effective technique to detect online abusive and bullying messages by merging natural language processing and machine learning. Two distinct freatures, namely Bag-of Words (BoW) and term frequency-inverse text frequency (TFIDF), are used to analyse the accuracy level of four distinct machine learning algorithms.","PeriodicalId":445708,"journal":{"name":"2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125319397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}