检索结果-内蒙古大学图书馆

The Effect of Several Kernel Functions to Financial Transaction Anomaly Detection Performance using One-Class SVM

学校读者我要写书评

暂无评论

The Effect of Several Kernel Functions to Financial Transact...

2019 International Congress on Applied information technology, AIT 2019

作者： Heryadi, Yaya Dandalina, D. Bina Nusantara University Computer Science Department Jakarta11530 Indonesia Master of Information and Technology Program Bina Nusantara University Jakarta11480 Indonesia

ISBN: (纸本)9781728158624

Anomalies in data become ubiquitous and often unavoidable. Despite it might be caused by various errors during the data collection or transportation, some anomalies potentially give important information such as indication of a new underlying process. The aim of anomaly detection task is to determine all such instances in the input dataset in a data-driven fashion. In the past decade, the proliferation of anomaly data in many domains have raised research interest resulted in a plethora of anomaly detection methods A vast number of previous study reports suggested that there is no single model will achieve the best performance for every dataset. This paper presents empiric results on the effect of several kernel functions to performance of One-Class Support Vector Machine (OC-SVM) as an anomaly detector model. The proposed model is tested using financial transactions of microfinance service dataset from an Indonesian Bank. The empiric results showed that OC-SVM model with Sigmoid and RBF kernels achieve the best statistically significant value of training, validation, and testing accuracies than the OC-SVM model with no-kernel, linear kernel and polynomial (degree 3,4,5,6) kernels. © 2019 IEEE.

关键词： Anomaly detection

AutoGMap: Learning to Map Large-scale Sparse Graphs on Memristive Crossbars

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Lyu, Bo Wang, Shengbo Wen, Shiping Shi, Kaibo Yang, Yin Zeng, Lingfang Huang, Tingwen Zhejiang Lab Hangzhou311121 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China Australian AI Institute Faculty of Engineering and Information Technology University of Technology SydneyNSW2007 Australia School of Information Science and Engineering Chengdu University Chengdu611040 China College of Science and Engineering Hamad Bin Khalifa University Doha5855 Qatar Science Program Texas A & M University at Qatar Doha23874 Qatar

The sparse representation of graphs has shown great potential for accelerating the computation of graph applications (e.g., Social Networks, Knowledge Graphs) on traditional computing architectures (CPU, GPU, or TPU). But the exploration of large-scale sparse graph computing on processing-in-memory (PIM) platforms (typically with memristive crossbars) is still in its infancy. To implement the computation or storage of large-scale or batch graphs on memristive crossbars, a natural assumption is that a large-scale crossbar is demanded, but with low utilization. Some recent works question this assumption, to avoid the waste of storage and computational resource, the fixed-size or progressively scheduled "block partition" schemes are proposed. However, these methods are coarse-grained or static, and are not effectively sparsity-aware. This work proposes the dynamic sparsity-aware mapping scheme generating method that models the problem with a sequential decision-making model, and optimizes it by reinforcement learning (RL) algorithm (REINFORCE). Our generating model (LSTM, combined with the dynamic-fill scheme) generates remarkable mapping performance on a small-scale graph/matrix data (complete mapping costs 43% area of the original matrix) and two large-scale matrix data (costing 22.5% area on qh882 and 17.1% area on qh1484). Our method may be extended to sparse graph computing on other PIM architectures, not limited to the memristive device-based platforms. Copyright © 2021, The Authors. All rights reserved.

关键词： Memory architecture

Implementation of Generate and Test Algorithm for Junior High School Zoning System in Malang 4

学校读者我要写书评

暂无评论

Implementation of Generate and Test Algorithm for Junior Hig...

4th International Conference on Vocational Education and Training, ICOVET 2020

作者： Ratnasar, Novia Agustin Mawarni, Inggar Tri Nur Handayani, Anik Wibawa, Aji Prasetya Hammad, Jehad A. H Wibowotomo, Budi Sugandi, MacHmud Universitas Negeri Malang Vocational Education Program Department Indonesia Universitas Negeri Malang Electrical Engineering Dept Indonesia Al-Quds Open University Faculty of Technology and Applied Sciences Department of Computer Information Systems Palestine Universitas Negeri Malang Industrial Technology Department Malang65145 Indonesia Universitas Negeri Malang Mechanical ENgineering Department Malang65145 Indonesia

ISBN: (纸本)9781728181318

Admission of new students in 2019 through zoning still faces many obstacles, one of which is the readiness of the application to determine the distance of a student's house from the nearest recommended school. The purpose of this study is to provide better recommendations regarding the shortest and most accurate distance in the zoning system for admission of junior high school students in Malang by using the Generate and Test algorithm without having to do it manually or using other map applications. The results of this study indicate that the distance value obtained from the haversine formula corresponds to the original distance measured through a zoning application that applies the Generate and Test algorithm to measure the closest distance from the student's location. The recommended distance is given through a process of repetition and comparison of nodes. By using the generate and test algorithm for the search process the shortest distance from the user to the destination school has an average time of 6.2MB, the level of accuracy is the same as the Haversine calculation. © 2020 IEEE.

关键词： Students

REVOLUTIONIZING RADIOTHERAPY COMMUNICATION SKILLS TRAINING: AN INNOVATIVE CHATBOT-BASED PROTOTYPE APPROACH (SCIMORT)

学校读者我要写书评

暂无评论

Journal of Medical Imaging and Radiation Sciences 2023年第3期54卷 S14-S15页

作者： Aniza, Azmi Nor Afiqah, Mohd Sukri Nur Ikhwan, Mustafa Kamal Muhamad Rozilawati, Ahmad Azmie, Yahya Noorazrul Safwan, Ahmad Fadzil Muhammad Adilah, Suhaimi Nani Noraini, Ahmad Wahid Izzuan, Hussin Saiful Liyana, Shuib Diagnostic Imaging and Radiotherapy Program Faculty of Health Sciences Universiti Kebangsaan Malaysia 50300 Kuala Lumpur Malaysia Department of Mathematical Sciences Faculty of Science and Technology Universiti Kebangsaan Malaysia 43600 Selangor Darul Ehsan Malaysia Department of Information Systems Faculty of Computer Science and Information Technology University of Malaya 50603 Kuala Lumpur Malaysia

OBJECTIVEThis study aimed to develop and evaluate the effectiveness of a chatbot-based prototype model called SCIMORT, designed to improve the communication skills of undergraduate students in Diagnostic Imaging and Radiotherapy at Universiti Kebangsaan Malaysia. MATERIALS & METHODSThis pilot and survey study involved the development of the SCIMORT module, which included the creation of a 3D virtual patient with visual and auditory cues. The SCIMORT module was evaluated by surveying 28 participants from the PDR cohorts. The questionnaire was designed to assess user acceptance and engagement of SCIMORT, and the expert validation of the questionnaire was assessed using Content Validity Ratio and Cronbach Alpha Coefficient. RESULTSA total of 28 participants from the PDR cohorts were enrolled in these projects. In the first phase, the virtual patient developed for SCIMORT exhibited visual cues such as blinking and eye movement, as well as auditory cues, including verbal responses and intonation of voice. During the second phase, expert validation of the questionnaire yielded favourable RESULTS: , with a Content Validity Ratio exceeding 83% and a Cronbach Alpha Coefficient surpassing 70%. However, both the control group (users' acceptance = 3.919 ± 1.245; users' engagement = 3.526 ± 1.270) and the Trial group (users' acceptance = 3.953 ± 1.114; users' engagement = 3.568 ± 1.142) showed neutral outcomes. CONCLUSIONThe SCIMORT module proved to be an effective tool for pre-clinical learning of communication skills in radiotherapy. However, the control and the Trial groups exhibited neutral results regarding users' acceptance and engagement. Further improvements are needed to make the module more interactive and adaptable to student needs.

关键词：

Deep convolutional neural network model for short-term electricity price forecasting

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Cheng, Hsu-Yung Kuo, Ping-Huan Shen, Yamin Huang, Chiou-Jye Department of Computer Science and Information Engineering National Central University Computer and Intelligent Robot Program for Bachelor Degree National Pingtung University School of Electrical Engineering and Automation Jiangxi University of Science and Technology

In the modern power market, electricity trading is an extremely competitive industry. More accurate price forecast is crucial to help electricity producers and traders make better decisions. In this paper, a novel method of convolutional neural network (CNN) is proposed to rapidly provide hourly forecasting in the energy market. To improve prediction accuracy, we divide the annual electricity price data into four categories by seasons and conduct training and forecasting for each category respectively. By comparing the proposed method with other existing methods, we find that the proposed model has achieved outstanding results, the mean absolute percentage error (MAPE) and root mean square error (RMSE) for each category are about 5.5% and 3, respectively. Copyright © 2020, The Authors. All rights reserved.

关键词： Power markets

Interactive 3D Segmentation for Primary Gross Tumor Volume in Oropharyngeal Cancer

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Saukkoriipi, Mikko Sahlsten, Jaakko Jaskari, Joel Orsmaa, Lotta Kangas, Jari Rasouli, Nastaran Raisamo, Roope Hirvonen, Jussi Mehtonen, Helena Järnstedt, Jorma Mäkitie, Antti Naser, Mohamed Fuller, Clifton Kann, Benjamin Kaski, Kimmo Department of Computer Science Aalto University School of Science Espoo Finland Faculty of Information Technology and Communication Sciences Computing Sciences University of Tampere Tampere Finland Department of Radiology Tampere University Faculty of Medicine and Health Technology Tampere University Hospital Tampere Finland Department of Otorhinolaryngology-Head and Neck Surgery Research Program in Systems Oncology Faculty of Medicine University of Helsinki Helsinki University Hospital Helsinki Finland Department of Radiation Oncology The University of Texas MD Anderson Cancer Center HoustonTX United States Artificial Intelligence in Medicine Program Mass General Brigham Harvard Medical School BostonMA United States Department of Radiation Oncology Dana-Farber Cancer Institute Brigham and Women's Hospital Harvard Medical School BostonMA United States The Alan Turing Institute British Library 96 Euston Rd LondonNW1 2DB United Kingdom

The main treatment modality for oropharyngeal cancer (OPC) is radiotherapy, where accurate segmentation of the primary gross tumor volume (GTVp) is essential. However, accurate GTVp segmentation is challenging due to significant interobserver variability and the time-consuming nature of manual annotation, while fully automated methods can occasionally fail. An interactive deep learning (DL) model offers the advantage of automatic high-performance segmentation with the flexibility for user correction when necessary. In this study, we examine interactive DL for GTVp segmentation in OPC. We implement state-of-the-art algorithms and propose a novel two-stage Interactive Click Refinement (2S-ICR) framework. Using the 2021 HEad and neCK TumOR (HECKTOR) dataset for development and an external dataset from The University of Texas MD Anderson Cancer Center for evaluation, the 2S-ICR framework achieves a Dice similarity coefficient of 0.713 ± 0.152 without user interaction and 0.824 ± 0.099 after five interactions, outperforming existing methods in both cases. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

Location embedding based pairwise distance learning for fine-grained diagnosis of urinary stones

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jin, Qiangguo Huang, Jiapeng Sun, Changming Cui, Hui Xuan, Ping Su, Ran Wei, Leyi Wu, Yu-Jie Wu, Chia-An Duh, Henry B.L. Lu, Yueh-Hsun School of Software Northwestern Polytechnical University Shaanxi China Yangtze River Delta Research Institute of NPU Suzhou China CSIRO Data61 Sydney Australia Department of Computer Science and Information Technology La Trobe University Melbourne Australia School of Engineering Shantou University Guangdong China College of Intelligence and Computing Tianjin University Tianjin China School of Information Xiamen University Xiamen China AIDD Faculty of Applied Science Macao Polytechnic University China Department of Radiology Shuang-Ho Hospital Taipei Medical University Taipei Taiwan Master of Public Health Program National Yang Ming Chiao Tung University Taipei Taiwan School of Design The Hong Kong Polytechnic University Hong Kong Department of Radiology School of Medicine College of Medicine Taipei Medical University Taipei Taiwan

The precise diagnosis of urinary stones is crucial for devising effective treatment strategies. The diagnostic process, however, is often complicated by the low contrast between stones and surrounding tissues, as well as the variability in stone locations across different patients. To address this issue, we propose a novel location embedding based pairwise distance learning network (LEPD-Net) that leverages low-dose abdominal X-ray imaging combined with location information for the fine-grained diagnosis of urinary stones. LEPD-Net enhances the representation of stone-related features through context-aware region enhancement, incorporates critical location knowledge via stone location embedding, and achieves recognition of fine-grained objects with our innovative fine-grained pairwise distance learning. Additionally, we have established an in-house dataset on urinary tract stones to demonstrate the effectiveness of our proposed approach. Comprehensive experiments conducted on this dataset reveal that our framework significantly surpasses existing state-of-the-art methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Embeddings

SkinSource: A Data-Driven Toolbox for Predicting Touch-Elicited Vibrations in the Upper Limb

学校读者我要写书评

暂无评论

SkinSource: A Data-Driven Toolbox for Predicting Touch-Elici...

International Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems (HAPTICS)

作者： Neeli Tummala Gregory Reardon Simone Fani Dustin Goetz Matteo Bianchi Yon Visell the Department of Electrical and Computer Engineering University of California Santa Barbara Santa Barbara CA USA the Media Arts and Technology Program University of California Santa Barbara Santa Barbara CA USA the School of Biological and Health Systems Engineering Arizona State University Tempe AZ USA the Soft Robotics for Human Cooperation and Rehabilitation Lab Instituto Italiano di Tecnologia Geneva GE Italy the Department of Mechanical Engineering University of California Santa Barbara Santa Barbara CA USA the Department of Information Engineering University of Pisa PI Italy the Department of Biological Engineering University of California Santa Barbara Santa Barbara CA USA

Vibrations transmitted throughout the hand and arm during touch contact play a central role in haptic science and engineering but are challenging to model or experimentally characterize. Here, we present SkinSource, a data-driven toolbox for predicting skin vibrations across the upper limb in response to user-specified input forces. The toolbox leverages impulse response measurements that encode the physics of vibration transmission across the hands and arms of four participants and provides software tools for analyzing the predicted skin responses. We show that the SkinSource predictions closely match experimental measurements and confirm the underlying assumption of linear vibration transmission in the skin. We also demonstrate through several usage examples how SkinSource can act as a versatile computational platform for haptic research applications, such as characterizing vibrotactile transmission in the skin, engineering haptic interfaces, and investigating touch perception.

关键词：

Find the Best Rule of Time Series Data Mining with Cluster Analysis

学校读者我要写书评

暂无评论

Journal of Physics: Conference Series 2021年第1期1830卷

作者： Relita Buaton Muhammad Zarlis Herman Mawengkang Syahril Efendi Graduate Program Of Computer Science Department Of Computer Science Faculty of Computer Science and Information Technology Universitas Sumatera Utara Medan Indonesia

Data mining is an analytical process of knowledge discovery in large and complex data sets. Many studies wish to explore data, to find information so that knowledge can be obtained through the grouping process, classification, rules discovery, associations and data mining visualization which shows similarity. Periodic data often occurs in business applications and sciences that has big size, high dimension and continuously updated. The similarity in periodic data is based on several approaches. One of common approaches is to transform periodic series into other domains so that dimensions are reduced, followed by index mechanism. Many studies of time series do not give optimal result because limited to extracting data not able to represent time series and its pattern which is then change into rules. Rules can be found in time series data, but they are still constrained by over fitting and difficult to present. It causes time series data and non linier function of data mining decision can't be optimal. The basic idea in the method proposed is to do periodic discretization for sub-sequential formation. These sub-sequences are grouped through a measure of similarity. The simple rule-finding technique is applied to obtain hidden rules in the temporal pattern. The optimal time series data expected to generate the uncertainty trend, previously unknown and can be used to make decisions or forecasting in the future.

关键词：