首页--工业技术论文--自动化技术、计算机技术论文--计算技术、计算机技术论文--计算机的应用论文--计算机网络论文--一般性问题论文

A Supervised Road Traffic Accidents Detection with Location Extraction from Microblogs

Acknowledgements第4-5页
Abstract第5页
摘要第6-11页
List of Abbreviations第11-14页
1. INTRODUCTION第14-19页
    1.1 Introduction第14-15页
    1.2 Motivation第15-16页
    1.3 Problem Statement第16页
    1.4 Research Questions第16-17页
    1.5 Significance of Study第17-18页
    1.6 Thesis Outline第18-19页
2. BACKGROUND AND RELATED WORK第19-35页
    2.1 Introduction第19页
    2.2 An Overview of Text Mining第19-23页
        2.2.1 Text Mining Processes第20-21页
        2.2.2 Data mining (DM) Techniques for Text第21-22页
        2.2.3 Information Extraction (IE)第22-23页
    2.3 The Social Media第23-25页
        2.3.1 Microblogs第23-24页
        2.3.2 Twitter第24-25页
    2.4 Related Works第25-33页
        2.4.1 Introduction to Event Detection第25-26页
        2.4.2 Characterization of Event Detection as Observed on Microblogs第26-27页
        2.4.3 Large Scale Event Detection from Microblogs第27-29页
        2.4.4 Small Scale Events Detection from Microblogs第29-31页
        2.4.5 Information Extraction from Microblogs第31-33页
    2.5 Detecting Road Traffic accidents from Microblogs第33-34页
    2.6 Summary第34-35页
3. METHODOLOGY第35-66页
    3.1 Introduction第35页
    3.2 Conceptual Framework第35-36页
    3.3 Data Acquisition第36-42页
        3.3.1 Data Collection using Twitter Streaming and Search API第37页
        3.3.2 Query Keywords Filtering第37-38页
        3.3.3 Study Area第38-41页
        3.3.4 Dataset第41-42页
    3.4 Road Traffic Accidents vs Non-road Traffic Classification第42-55页
        3.4.1 Preprocessing Phase 1: Cleaning Tweets and Filtering第43-44页
        3.4.2 Preprocessing Phase 2: Feature Vectors Extraction第44-46页
        3.4.3 Training the SVM Model第46-48页
        3.4.4 Experiments and Results第48-49页
        3.4.5 Evaluation of Classifier第49-54页
        3.4.6 Discussion第54-55页
    3.5 Traffic Accident Location Extraction第55-64页
        3.5.1 Introduction to Named Entity Extraction第55-56页
        3.5.2 Syntactic Analysis第56-58页
        3.5.3 Proposed Noun Phrase with N-gram Pattern Matching for Location Extraction第58-60页
        3.5.4 Noun Phrase Extraction第60-61页
        3.5.5 Filters: Preposition and Regular Expressions第61页
        3.5.6 N-gram Location Matching第61-62页
        3.5.7 Geographical Name Matching第62-63页
        3.5.8 Experiment and Results第63-64页
    3.6 Summary第64-66页
4.WEB APPLICATION PROTOTYPE FOR TRAFFIC ACCIDENT DETECTION SYSTEM第66-81页
    4.1 Introduction第66页
    4.2 Requirement Analysis第66-69页
        4.2.1 Functional Requirements for the Traffic Accident Detection System (TADS)第66-67页
        4.2.2 Use cases Diagrams第67页
        4.2.3 Non-functional Requirements第67-68页
        4.2.4 System Environment Requirements第68-69页
        4.2.5 Constraints第69页
    4.3 System Design第69-74页
        4.3.0 Activity Diagrams第69-71页
        4.3.1 Sequence Diagram第71-72页
        4.3.2 Class Diagram第72-73页
        4.3.3 Database Design第73-74页
    4.4 Implementation第74-78页
        4.4.1 Backend Implementation第75-77页
        4.4.2 Front End第77-78页
    4.5 System Testing第78-80页
    4.6 Summary第80-81页
5. CONCLUSION AND FUTURE WORK第81-85页
References第85-90页
Appendices第90-102页
    Appendix A: Tweet JSON Metadata第90-92页
    Appendix B: Sample Codes: Back End第92-98页
    Appendix C: Sample Codes: Front End第98-101页
    Appendix D: Snapshots of TADS第101-102页

论文共102页,点击 下载论文
上一篇:基于OBDD的工控网络连通可靠性分析
下一篇:Gamified Web Application System for Court Training