Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence最新文献

Using Web Pages Dynamicity to Prioritise Web Crawling 使用网页动态来优先考虑网络爬虫

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366757

Nisreen Alderratia, M. Elsheh

{"title":"Using Web Pages Dynamicity to Prioritise Web Crawling","authors":"Nisreen Alderratia, M. Elsheh","doi":"10.1145/3366750.3366757","DOIUrl":"https://doi.org/10.1145/3366750.3366757","url":null,"abstract":"Web crawling is a process performed to collect web pages from the web, in order to be indexed and used for displaying the search results according to users' requirements. In addition, web crawlers must continually revisit web pages, to keep the search engine database updated. Moreover, it is fundamental to determine in the crawling process, the most important pages to be recrawled first. This is to avoid the time limitation and network issues that face the web crawling process. Thus, this research attempts to introduce a method that is used to indicate the crawler, specifically, in order to identify in what order it should recrawl web pages that have been crawled before, as to acquire more important and valuable pages earlier than others. In addition, the researchers proposed a web crawling strategy which is based on the topic similarity, accompanied with the dynamicity of web pages, where the crawler was downloading relevant pages and recrawling them recursively. Also, every time a change emerged in one of the pages, its counter increased. Therefore, if the page was relevant and changed frequently it would be considered an important page and was given a high priority in the crawling process. The obtained results indicated that using web pages' dynamicity is an effective way for prioritising web pages in the crawling process, in order to obtain the highest dynamic pages first, as there is a high possibility of being changed in terms of their content, before the least dynamic ones.","PeriodicalId":145378,"journal":{"name":"Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116140226","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}

引用次数: 2

NMF-based DCG Optimization for Collaborative Ranking on Recommendation Systems 基于nmf的推荐系统协同排序的DCG优化

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366753

Noor Ifada, Dziyaur Rohman Miftah Alim, M. K. Sophan

引用次数: 1

Using Distillation to Improve Network Performance after Pruning and Quantization 利用精馏提高网络在修剪和量化后的性能

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366751

Zhenshan Bao, Jiayang Liu, Wenbo Zhang

{"title":"Using Distillation to Improve Network Performance after Pruning and Quantization","authors":"Zhenshan Bao, Jiayang Liu, Wenbo Zhang","doi":"10.1145/3366750.3366751","DOIUrl":"https://doi.org/10.1145/3366750.3366751","url":null,"abstract":"As the complexity of processing issues increases, deep neural networks require more computing and storage resources. At the same time, the researchers found that the deep neural network contains a lot of redundancy, causing unnecessary waste, and the network model needs to be further optimized. Based on the above ideas, researchers have turned their attention to building more compact and efficient models in recent years, so that deep neural networks can be better deployed on nodes with limited resources to enhance their intelligence. At present, the deep neural network model compression method have weight pruning, weight quantization, and knowledge distillation and so on, these three methods have their own characteristics, which are independent of each other and can be self-contained, and can be further optimized by effective combination. This paper will construct a deep neural network model compression framework based on weight pruning, weight quantization and knowledge distillation. Firstly, the model will be double coarse-grained compression with pruning and quantization, then the original network will be used as the teacher network to guide the compressed student network. Training is performed to improve the accuracy of the student network, thereby further accelerating and compressing the model to make the loss of accuracy smaller. The experimental results show that the combination of three algorithms can compress 80% FLOPs and reduce the accuracy by only 1%.","PeriodicalId":145378,"journal":{"name":"Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130874822","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}

引用次数: 3

Report on a Hackathon for Car Navigation Using Traffic Risk Data 关于利用交通风险数据进行汽车导航的黑客马拉松的报告

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366758

S. Ito, K. Zettsu

引用次数: 4

A Metaheuristic Algorithm for the Probabilistic Orienteering Problem 概率定向问题的一种元启发式算法

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366761

Xiaochen Chou, L. Gambardella, R. Montemanni

引用次数: 3

Content based Image Retrieval with Rocchio Algorithm for Relevance Feedback Using 2D Image Feature Representation 基于二维图像特征表示的Rocchio相关反馈算法的图像检索

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366755

I. Siradjuddin, Aryandi Triyanto, S. MochammadKautsar

引用次数: 3

ALT-Based Route Planning in Dynamic Time-Dependent Road Networks 动态时变路网中基于alt的路径规划

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366752

Famei He, Yin Xu, Xuren Wang, Anran Feng

引用次数: 2

Spam Detection Using Clustering-Based SVM 基于聚类支持向量机的垃圾邮件检测

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366754

Darshit Pandya

引用次数: 7

Hydrological Time Series Prediction Model Based on Attention-LSTM Neural Network 基于注意力- lstm神经网络的水文时间序列预测模型

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366756

Yiran Li, Juan Yang

{"title":"Hydrological Time Series Prediction Model Based on Attention-LSTM Neural Network","authors":"Yiran Li, Juan Yang","doi":"10.1145/3366750.3366756","DOIUrl":"https://doi.org/10.1145/3366750.3366756","url":null,"abstract":"Constructing predictive models with neural networks has always been the focus of research in chaotic time series prediction. Since Hinton proposed the concept of deep learning in 2006, the development of neural networks is getting faster and faster, and a variety of neural networks appear. Among them, RNN neural network and LSTM neural network are applied to all fields of our daily life for the property of well-processed time series. Therefore, we use LSTM neural network to construct our model, and optimize the model by Attention mechanism to establish an Attention-LSTM hydrological time series prediction model. The experimental results show that the Attention-LSTM model has better improvement in the mean square error and absolute error of the predicted values than the common LSTM model and the traditional BP model. And due to the introduction of the Attention mechanism, it can highlight the key factors to some extent. influences.The experimental results show that the Attention-LSTM model has the advantages of high prediction accuracy and small lag error, which is helpful for the application of deep learning algorithm in hydrological time series prediction.","PeriodicalId":145378,"journal":{"name":"Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125537484","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}

引用次数: 6

Understanding Usability and User Acceptance of Usage-Based Insurance from Users' View 从用户的角度理解基于使用的保险的可用性和用户接受度

Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence Pub Date : 2019-09-18 DOI: 10.1145/3366750.3366759

Juan Quintero, Z. Benenson

{"title":"Understanding Usability and User Acceptance of Usage-Based Insurance from Users' View","authors":"Juan Quintero, Z. Benenson","doi":"10.1145/3366750.3366759","DOIUrl":"https://doi.org/10.1145/3366750.3366759","url":null,"abstract":"Intelligent Transportation Systems (ITS) cover a variety of services related to topics such as traffic control and safe driving, among others. In the context of car insurance, a recent application for ITS is known as Usage-Based Insurance (UBI). UBI refers to car insurance policies that enable insurance companies to collect individual driving data using a telematics device. Collected data is analysed and used to offer individual discounts based on driving behaviour and to provide feedback on driving performance. Although there are plenty of advertising materials about the benefits of UBI, the user acceptance and the usability of UBI systems have not received research attention so far. To this end, we conduct two user studies: semi-structured interviews with UBI users and a qualitative analysis of 186 customer inquiries from a web forum of a German insurance company. We find that under certain circumstances, UBI provokes dangerous driving behaviour. These situations could be mitigated by making UBI transparent and the feedback customisable by drivers. Moreover, the country driving conditions, the policy conditions, and the perceived driving style influence UBI acceptance.","PeriodicalId":145378,"journal":{"name":"Proceedings of the 2019 2nd International Conference on Machine Learning and Machine Intelligence","volume":"223 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115618518","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}

引用次数: 2