检索结果-内蒙古大学图书馆

2nd international conference on Communications and Cyber-Physical Engineering (ICCCE)

作者： Reddy, V. Hemalatha Kumari, Soumya Muralidharan, Vinitha Gigoo, Karan Thakare, Bhushan S. STESs Sinhgad Acad Engn Pune Maharashtra India

ISBN: (纸本)9789811387159;9789811387142

Nowadays, with easy access to internet, food is delivered at our doorsteps just on the click of a button due to which people have started to consume higher amount of fast food. This has accelerated the chances of suffering from a chronic disease known as obesity. Since obesity has become such a widespread disease, various mobile e-health applications have been developed for assistive calorie measurement to help people fight against health-related problems. In this paper, we have surveyed different methods for food recognition and calorie measurement using various methods and compared their performances based on several factors.

关键词： Convolutional neural network (CNN) Deep learning Food recognition machine learning (ML)

来源：评论

学校读者我要写书评

暂无评论

Knowledge-graph based Proactive Dialogue Generation with Improved Meta-learning 2020

Knowledge-graph based Proactive Dialogue Generation with Imp...

引用

2nd international conference on Image processing and machine Vision / international conference on Pattern Recognition and machine learning (IPMV)

作者： Xu, Hongcai Bao, Junpeng Wang, Junqing Xi An Jiao Tong Univ Xian 710049 Shaanxi Peoples R China

ISBN: (纸本)9781450388412

Knowledge graph-based dialogue systems can narrow down knowledge candidates for generating informative and diverse responses with the use of prior information, e.g., triple attributes or graph paths. However, most current knowledge graphs (KG) cover incomplete domain-specific knowledge. To overcome this drawback, we propose a knowledge graph based proactive dialogue generation model (KgDg) with three components, improved model-agnostic meta-learning algorithm (MAML), knowledge selection in knowledge triplets embedding, and knowledge aware proactive response generator. For knowledge triplets embedding and selection, we formulate it as a problem of sentence embedding to better capture semantic information. Our improved MAML algorithm is capable of learning general features from a limited number of knowledge graphs, which can also quickly adapt to dialogue generation with unseen knowledge triplets. Extensive experiments are conducted on a knowledge aware dialogue dataset (DuConv). The results show that KgDg adapts both fast and well to knowledge graph based dialogue generation and outperforms state-of-the-art baselines.

关键词： Knowledge graph Proactive dialogue generation Meta-learning

来源：评论

学校读者我要写书评

暂无评论

A Forgotten Strategy for Pooled Contextualized Embedding learning 2020

A Forgotten Strategy for Pooled Contextualized Embedding Lea...

引用

2nd international conference on Image processing and machine Vision / international conference on Pattern Recognition and machine learning (IPMV)

作者： Cai, Dunbo Huang, Zhiguo Qian, Ling Hang, Tao Feng, Qingsong China Mobile Suzhou Software Suzhou Peoples R China

ISBN: (纸本)9781450388412

String (word) embeddings are keys to advanced neural natural language understanding (NLU) models. Recently, a new embedding called contextual string embedding (CtxEmb) reached a new state of the art on many NLU tasks. Specially, a pooling variant of CtxEmb demonstrated new best results on named entity recognition (NER) tasks. While the pooling variant is good at reaching new best performance in terms of Micro F1 score, it poses questions on what kind of information in the pool is helpful. In particular, the pooling variant maintains a (possibly long) list of previously seen occurrences of a word, and computes embedding for the word in a new context based on those occurrences. In this paper, we propose a strategy to forget some previous occurrences, and thus exploring which history are beneficial in constructing embeddings for a new context of a word. The proposed forgotten strategy is designed by accounting for a distance metric defined in this paper. Preliminary experiments on the WNUT-17 task show the effectiveness of our forgotten strategy, and uncover that embeddings that are diverse in terms of cosine similarity are helpful in forming an aggregated embedding.

关键词： Word Embedding learning Contextual String Embedding Representation learning

来源：评论

学校读者我要写书评

暂无评论

Atrial Fibrillation Pattern Recognition using Features of Second Order Dynamic System

Atrial Fibrillation Pattern Recognition using Features of Se...

引用

Intelligent Cybernetics Technology & Applications (ICICyTA), international conference on

作者： Lee Wei Qi Nurul Ashikin Abdul-Kadir Wei Wei Heng Mohd Afzan Othman Norlaili Mat Safri Abdul-Mutalib Embong Universiti Teknologi Malaysia Johor Bahru Malaysia University of Southampton Malaysia Johor Malaysia Universiti Malaysia Terengganu Terengganu Malaysia

ISBN: (纸本)9798350399141

According to World Health Organization (WHO), an estimated 17.9 million people died from cardiovascular diseases (CVD) in 2019, representing 32 % of all global deaths. Of these deaths, 85 % were due to heart attack and stroke. The occurrence and prevalence of atrial fibrillation (AF) is growing worldwide. Limited tools are available to evaluate clinical outcomes and response to thrombolysis in stroke patients with AF. Therefore, this study analysed the ECG features of AF and the normal sinus rhythm (NSR) signals for AF recognition. The first objective is to extract AF features using second-order dynamic system (SODS) algorithm. The following objective is to investigate the effect of windowing length towards AF classification. Next, to compare the two-pattern recognition machine learning support vector machine (SVM) and artificial neural network (ANN) on the accuracy, specificity, and sensitivity of AF classification. In this study, the Physiobank database, included MITBIH Atrial Fibrillation Dataset and MITBIH Normal Sinus Rhythm Dataset were used. For signal pre-processing, butterworth filter are used to diminish the muscle noise and the features signals were extracted. Multiple episodes of the windowing size of 2s, 4s, 6s, 8s and 10s were included in this design to evaluate the appropriate windowing size for AF recognition. The pattern recognition machine learning of SVM algorithm has higher accuracy compared to ANN, which are 100 % with 4s windowing size. In conclusion, the 4s windowing size having the highest detection rate in AF classification system.

关键词： Support vector machines machine learning algorithms Atrial fibrillation machine learning Artificial neural networks Electrocardiography Stroke (medical condition)

来源：评论

学校读者我要写书评

暂无评论

Ac operation hardware learning neural circuit using v-f converter system 2nd

Ac operation hardware learning neural circuit using v-f conv...

引用

2nd international conference on Sensor Networks and signal processing, SNSP 2019

作者： Kawaguchi, Masashi Ishii, Naohiro Umeno, Masayoshi National Institute of Technology Suzuka College Shiroko Mie Suzuka510-0294 Japan Aichi Institute of Technology Yachigusa Yagusa-cho Aichi Toyota470-0392 Japan Chubu University 1200 Matsumoto-cho Aichi Kasugai487-8501 Japan

ISBN: (纸本)9789811549168

In the artificial intelligence, machine learning and neural network field, many practical use models have been proposed. However, these models are based on a digital Von Neumann model’s computer. There are minority studies in the field of analog learning neural networks. Early analog hardware neural network models were configured with the operational amplifier and the resistance element. It is difficult to change the value of solid resistance for the learning process. In the present paper, proposed is a learning neural network using electronic alternating current (AC) circuits including a voltage-frequency (V-F) converter. These circuits are composed of an amplifier, additional circuit, inverter, subtract circuit, rectifier circuit and V-F converter. The input voltage value was described as the input signal and the input frequency considered the connecting weights. It is easy to change the connecting weights with a V-F converter. Finally, the input frequency converges to a constant value after only several learning process. The learning count time is extremely small. The learning time is quite fast in this AC transmission circuit. The model works using pure analog electronic circuits. The learning time is quite short compare to with a digital process computer. © Springer Nature Singapore Pte Ltd 2021.

关键词： learning systems

来源：评论

学校读者我要写书评

暂无评论

Extreme learning machine Based Defect Detection for Solder Joints

引用

JOURNAL OF INTERNET TECHNOLOGY 2020年第5期21卷 1535-1543页

作者： Ma, Liyong Xie, Wei Zhang, Yong Feng, Xijia Harbin Inst Technol Sch Informat Sci & Engn Harbin Heilongjiang Peoples R China Harbin Inst Technol Dept Civil Engn Harbin Heilongjiang Peoples R China

The quality of solder joints is essential for electronic products, and the detection of defects in solder joints is critical to the quality control of electronic products. A vision inspection is developed to detect defects of solder joints in automatic line. Extreme learning machine is applied to identify defective solder joints from qualified ones. Five low level features and three advanced features are employed as input features. The low-level features include roundness, roughness, entropy, contrast and histogram of oriented gradient. The advanced features include grey-level co-occurrence matrix, local binary pattern, and segmentation-based fractal texture analysis. To solve unbalanced samples problem, Gaussian mixture model based dense estimation scheme is proposed to adjust the classification super plane for extreme learning machine. The experimental results demonstrate that the proposed defect detection method is more efficient than neural network, support vector machines, common extreme learning machine and convolutional neural network-based methods, and it has real-time performance to meet the equirement of the actual production line.

关键词： Defect detection Solder joints Extreme learning machine Gaussian mixture model

来源：评论

学校读者我要写书评

暂无评论

Research on Fault Detection Method of Catenary Equipment Based on Deep learning 2

Research on Fault Detection Method of Catenary Equipment Bas...

引用

2nd IEEE international conference on Information Technology, Big Data and Artificial Intelligence, ICIBA 2021

作者： Zhang, Fangheng Tao, Kemeng Xie, Xiaoxia Liu, Hang Tian, Lijun Yang, Xixi Wang, Minghan Heyuan Intelligent Technology CO. LTD Jinan China Jinan Union IT Information Technology CO. LTD Jinan China School of Electrical Engineering Shandong University Jinan China

ISBN: (纸本)9781665428767

The transportation capacity of electrified railways can be more than double that of ordinary railways, and it has the advantages of low cost, low energy consumption and low environmental pollution. As one of the core and most important components of the electrified railway, the condition of the catenary will directly affect the safety and stable operation of railways. Therefore, the research of catenary fault detection and identification technology is very important. Aiming at the inefficiency of traditional methods, this paper proposes a deep learning-based catenary fault detection technology, introduces image processing technology into the electrified railway catenary detection system, and processes the photographs of railway equipment to realize the catenary equipment fault detection. Experiments prove that this method has good performance. © 2021 IEEE.

关键词： Fault detection

来源：评论

学校读者我要写书评

暂无评论

Random Forest-Based Enhanced Detection of Lumpy Skin Disease Under Two Data Types

Random Forest-Based Enhanced Detection of Lumpy Skin Disease...

引用

international conference on signal processing, Communication, Power and Embedded System (SCOPES)

作者： Madhumita Pal Ganapati Panda Rajiv Lochan Bal Dept. of Electrical Engineering Government College of Engineering Keonjhar Odisha India Dept. of Electronics & Telecommunication Engineering C.V.Raman Global University Bhubaneswar India Dept. of Mining Engineering Orissa School of Mining Engg. Keonjhar India

ISBN: (数字)9798331506452

ISBN: (纸本)9798331506469

The deadly and extremely contagious Lumpy Skin Lesion Disease (LSLD) frequently affects cattle. Tough raised nodules that form on the cattle's body, high fever, sterility, and a marked reduction in milk output by nursing cattle are typical symptoms of LSLD. Drooling eyes, increased nasal and salivary secretions, genital and limb problems, and a reluctance to move or eat are among the symptoms. Furthermore, predicting the course of the disease in advance can significantly lessen the economic burden, even though the mortality rate among cattle infected with LSLD is low. A random forest (RF) machine learning (ML) model was used in this work to ascertain whether or not cattle have lumpy skin disease. The paper's main goal is to use an RF model to discover and classify lumpy skin lesions in cattle as normal, mild, or severe. Veterinarians can save many cattle's lives by detecting the sickness early. To evaluate the effectiveness of the random forest model, we also looked at two aspects of the dataset: balanced and imbalanced. The random forest model with an unbalanced dataset yielded the following results: 0.74, 0.76, 0.89, 0.82, 0.63&0.75 for normal skin, 0.74, 0.52, 0.35, 0.42, 0.63, and 0.33 for mild lumpy skin lesions, and 0.74,0.50,0.22,0.31,0.60, and 0.18 for severe lumpy skin lesions. These findings were derived for the precision-recall curve (P-R), accuracy (Arc), precision (Pn), recall (RI), Fl-score, and the area under the curve (AUC). Accuracy (Arc) 97.13, precision (Pn) 0.96, recall (RI) 0.91, F1-score 0.93, the area under the curve (AUC) 0.94, and P-R curve value 0.90 are the metrics for categorizing normal skin for a balanced dataset. The following values were obtained to classify minor lumpy skin lesions: Arc 97.13, Pn 0.92, RI 0.94, F1-score 0.93, AUC 0.95, and P-R 0.88. An Arc of 97.13, pn of 0.96, Rl of 0.99, F1-score of 0.97, AUC value of 0.98, and P-R value of 0.95 are the results for severe lumpy skin lesions. Various training schemes can also evalua

关键词： Radio frequency Training Accuracy signal processing algorithms Cows signal processing Skin Lesions Random forests Diseases

来源：评论

学校读者我要写书评

暂无评论

Detecting domain name system tunneling and exfiltration from domain name system traffic 2nd

Detecting domain name system tunneling and exfiltration from...

引用

2nd international conference on Sensor Networks and signal processing, SNSP 2019

作者： Tseng, Yi-Chung Sun, Ming-Kung Chen, Wei-An Acer Cyber Security Incorporation Taipei Taiwan

ISBN: (纸本)9789811549168

In this study, we present a framework to detect a Domain Name System (DNS) tunnel and DNS exfiltration network traffic by using both unsupervised and supervised learning algorithms. In general, considerable time is required to learn the structure of the data before clustering when using an unsupervised learning algorithm. Therefore, in this study, we leveraged the power of mathematical algorithms for calculating the optimal number of clusters and reducing the time required for understanding the data structure. Conversely, we used a supervised learning method to learn the data leakage behavior for detecting DNS exfiltration traffic. We used an open-source tool to generate testing data, and the experimental result proved the robustness of the proposed framework. © Springer Nature Singapore Pte Ltd 2021.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Application and optimization of intrusion detection system based on Artificial Intelligence 2

Application and optimization of intrusion detection system b...

引用

2nd IEEE international conference on Civil Aviation Safety and Information Technology (ICCASIT)

作者： Zhu, Kaidi Wu, Xiangwen Yu, Sun Huo, Da Li, Rui Pei, Xikai Second Res Inst CAAC Chengdu Peoples R China Inner Mongolia Autonomous Reg Civil Airports Grp Hulun Buir Branch Hulunbeir Peoples R China North China Regionnal Air Traff Management Bur CA Beijing Peoples R China China Civilaviat Adm Air Traff Management Bur Beijing Peoples R China

ISBN: (纸本)9781728199481

The traditional perimeter intrusion detection system will be faced more challenges in the new era, such as high-accuracy positioning, low energy dependence, high environmental robustness and so on. This paper introduces a new type of artificial intelligence intrusion detection technology. The RF sensor chip is embedded in the detection cable, and the radio frequency electromagnetic environment is formed in space through the sending and receiving of radio frequency signal, which is used to sense the surrounding environment to detect electromagnetic *** advanced space-time signal processing technology and machine learning technology, the influence of environmental interference on the system is effectively eliminated, and the intrusion detection system based on artificial intelligence algorithm has lower false alarm rate and higher environmental robustness.

关键词： image processing machine learning Intrusion Detection Technology Microwave array technology

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：