Soft Computing in Big Data Processing [recurso electrónico

La 4è de couv. indique : "Big data is an essential key to build a smart world as a meaning of the streaming, continuous integration of large volume and high velocity data covering from all sources to final destinations. The big data range from data mining, data analysis and decision making, by drawing statistical rules and mathematical patterns through systematical or automatically reasoning. The big data helps serve our life better, clarify our future and deliver greater value. We can discover how to capture and analyze data. Readers will be guided to processing system integrity and implementing intelligent systems. With intelligent systems, we deal with the fundamental data management and visualization challenges in effective management of dynamic and large-scale data, and efficient processing of real-time and spatio-temporal data. Advanced intelligent systems have led to managing the data monitoring, data processing and decision-making in realistic and effective way. Considering a big size of data, variety of data and frequent changes of data, the intelligent systems basically challenge new data management tasks for integration, visualization, querying and analysis. Connected with powerful data analysis, the intelligent systems will provide a paradigm shift from conventional store and process systems. This book focuses on taking a full advantage of big data and intelligent systems processing. It consists of 11 contributions that feature extraction of minority opinion, method for reusing an application, assessment of scientific and innovative projects, multi-voxel pattern analysis, exploiting No-SQL DB, materialized view, TF-IDF criterion, latent Dirichlet allocation, technology forecasting, small world network, and classification & regression tree structure. This edition is published in original, peer reviewed contributions covering from initial design to final prototypes and authorization." Soft Computing in Big Data Processing Preface 6 Contents 9 Extraction of Minority Opinion Based on Peculiarity in a Semantic Space Constructed of Free Writing 11 1 Introduction 11 2 Technical Elements of the Proposed Method 12 2.1 Morphological Analysis 12 2.2 Latent Semantic Indexing 13 3 ProposedMethod 13 3.1 Basic Policy of the Proposal 13 3.2 Extraction of Minority Opinion Using the Peculiarity Factor 14 3.3 Proposed Method Procedure 15 4 Experiment 15 4.1 Experimental Settings 15 4.2 Experimental Result 16 4.3 Discussion 17 5 Conclusions 19 References 19 System and Its Method for Reusing an Application 21 1 Introduction 21 2 The Trends of Related Research 22 2.1 Introduction to the Open USN Service Platform 22 2.2 Functional Architecture of the Open USN Service Framework 23 3 The Design of the Pre-registered Application Based Reuse System[4] 25 3.1 The Processing Procedure of the Pre-Registered Application Based Reuse System 26 4 Conclusions 28 References 28 Assessment of Scientific and Innovative Projects: Mathematical Methods and Models for Management Decisions 29 1 Introduction 29 2 Materials and Methods 30 3 Experimental Results 34 4 Discussion and Conclusion 37 References 38 Multi-Voxel Pattern Analysis of fMRI Based on Deep Learning Methods 39 1 Introduction 39 2 Decoding Process Based on DBN 40 2.1 Restricted Boltzmann Machine 40 2.2 Deep Brief Network 42 2.3 Decoding Process for fMRI Data 43 3 Decoding Experiments 44 3.1 Experiment Conditions 44 3.2 Decoding Results 45 3.3 Discussion 46 4 Conclusion 47 References 47 Exploiting No-SQL DB for Implementing Lifelog Mashup Platform 49 1 LifelogMashup 49 1.1 Lifelog Services and Mashups 49 2 Previous Work 50 2.1 Lifelog Mashups Platform 50 2.2 Lifelog Common Data Model 51 2.3 Lifelog API 51 2.4 Limitations in Previous Work 52 2.5 Research Goal and Approach 52 3 Exploiting No-SQL DB for Implementing Lifelog Mashup Platform 53 3.1 Using MongoDB for Lifelog Mashup 53 3.2 Implementing LLCDM with MongoDB 54 3.3 Implementing LLAPI with MongoDB 55 4 Case Study 57 5 Conclusion 58 References 58 Using Materialized View as a Service of Scallop4SC for Smart City Application Services 60 1 Smart City and Scallop4SC 60 1.1 Smart City and Services 60 1.2 Scallop4SC (Scalable Logging Platform for Smart City) 61 1.3 IntroducingMaterialized View in Scallop4SC 61 2 MaterializedView as a Service for Large-Scale House Log 63 2.1 Materialized View as a Service (MVaaS) 63 2.2 House Log Stored in Scallop4SC 63 2.3 Converting Raw Data into Materialized View 64 3 Case Study 65 3.1 Power Consumption Visualization Service 65 3.2 Wasteful Energy Detection Service 66 3.3 Diagnosis and Improvement of Lifestyle Support Service 67 4 Conclusion 68 References 69 Noise Removal Using TF-IDF Criterion for Extracting Patent Keyword 70 1 Introduction 70 2 Text Mining 71 3 TF-IDF 71 4 Social Network Analysis 72 5 Research Methodology 73 6 Experiment 75 7 Conclusion 77 References 78 Technology Analysis from Patent Data Using Latent Dirichlet Allocation 79 1 Introduction 79 2 Literature Review 80 3 Methodology 81 3.1 Data Preprocessing 81 3.2 The Tf-Idf 81 3.3 Latent Dirichlet Allocation (LDA) 82 3.4 Determining the Number of Topics 83 4 Experimental Results 84 5 Conclusion 87 References 87 A Novel Method for Technology Forecasting Based on Patent Documents 89 1 Introduction 89 2 Literature Review 90 2.1 Patent Information 90 2.2 Technology Forecasting 90 3 Emerging Technology Forecasting Model 91 3.1 Deriving a Patent–Keyword Matrix 91 3.2 Deriving a Principal Component–Patent Matrix by Applying PCA 92 3.3 Clustering Patent Document by Applying K-medoids Algorithm 92 3.4 Forecasting Emerging Technology Using Patent Information 93 4 Constructing and Verifying the Emerging Technology Model 93 5 Conclusion 97 References 98 A Small World Network for Technological Relationship in Patent Analysis 99 1 Introduction 99 2 Small World Network 100 3 Technological Relationship by Small World Network 100 4 Simulation and Experiment 102 5 Conclusions 107 References 107 Key IPC Codes Extraction Using Classification and Regression Tree Structure 108 1 Introduction 108 2 Research Background 109 3 A Classification and Regression Tree Model for IPC Data Analysis 109 4 Experimental Results 111 5 Conclusions 115 References 115 Author Index 117 La 4è de couv. indique : "Big data is an essential key to build a smart world as a meaning of the streaming, continuous integration of large volume and high velocity data covering from all sources to final destinations. The big data range from data mining, data analysis and decision making, by drawing statistical rules and mathematical patterns through systematical or automatically reasoning. The big data helps serve our life better, clarify our future and deliver greater value. We can discover how to capture and analyze data. Readers will be guided to processing system integrity and implementing intelligent systems. With intelligent systems, we deal with the fundamental data management and visualization challenges in effective management of dynamic and large-scale data, and efficient processing of real-time and spatio-temporal data. Advanced intelligent systems have led to managing the data monitoring, data processing and decision-making in realistic and effective way. Considering a big size of data, variety of data and frequent changes of data, the intelligent systems basically challenge new data management tasks for integration, visualization, querying and analysis. Connected with powerful data analysis, the intelligent systems will provide a paradigm shift from conventional store and process systems. This book focuses on taking a full advantage of big data and intelligent systems processing. It consists of 11 contributions that feature extraction of minority opinion, method for reusing an application, assessment of scientific and innovative projects, multi-voxel pattern analysis, exploiting No-SQL DB, materialized view, TF-IDF criterion, latent Dirichlet allocation, technology forecasting, small world network, and classification & regression tree structure. This edition is published in original, peer reviewed contributions covering from initial design to final prototypes and authorization." Front Matter....Pages 1-9 Extraction of Minority Opinion Based on Peculiarity in a Semantic Space Constructed of Free Writing....Pages 1-10 System and Its Method for Reusing an Application....Pages 11-18 Assessment of Scientific and Innovative Projects: Mathematical Methods and Models for Management Decisions....Pages 19-28 Multi-Voxel Pattern Analysis of fMRI Based on Deep Learning Methods....Pages 29-38 Exploiting No-SQL DB for Implementing Lifelog Mashup Platform....Pages 39-49 Using Materialized View as a Service of Scallop4SC for Smart City Application Services....Pages 51-60 Noise Removal Using TF-IDF Criterion for Extracting Patent Keyword....Pages 61-69 Technology Analysis from Patent Data Using Latent Dirichlet Allocation....Pages 71-80 A Novel Method for Technology Forecasting Based on Patent Documents....Pages 81-90 A Small World Network for Technological Relationship in Patent Analysis....Pages 91-99 Key IPC Codes Extraction Using Classification and Regression Tree Structure....Pages 101-109 Back Matter....Pages 111-111