Content: Machine generated contents note: 1.Introduction --
1.1.Overview of Knowledge Bases --
1.2.Overview of Knowledge Extraction in Knowledge Bases --
1.2.1.Extraction Techniques Overview --
1.2.2.Representation Models Overview --
1.3.Overview of Knowledge Bases Question Answering --
1.3.1.Question Answering in Curated KBs --
1.3.2.Question Answering in Open KBs --
1.4.Research Issues in Querying and Extracting Knowledge Bases --
1.4.1.An Architecture for Knowledge Base Management --
1.4.2.Our Contributions --
1.5.Book Organization --
2.Cache Based Optimization for Querying Curated Knowledge Bases --
2.1.Design Overview --
2.2.The SPARQL Endpoint Cache Framework --
2.2.1.Query Distance Calculation --
2.2.2.Feature Modelling --
2.2.3.Suggesting and Prefetching Similar Queries --
2.2.4.Caching and Replacement --
2.3.Experimental Evaluations and Discussions --
2.3.1.Setup --
2.3.2.Analysis of Real-World SPARQL Queries --
2.3.3.Performance of Cache Replacement Algorithm Note continued: 2.3.4.Comparison of Feature Modelling Approaches --
2.3.5.Performance Comparison with the State-of-the-Art --
2.3.6.Discussions --
2.4.Related Work --
2.4.1.Semantic Caching --
2.4.2.Query Suggestion --
2.5.Summary --
3.Query Performance Prediction on Knowledge Base --
3.1.Design Overview --
3.1.1.Motivation --
3.1.2.Challenges --
3.1.3.Prediction Approach Overview --
3.2.Preliminaries --
3.2.1.Multiple Regression --
3.2.2.Dimension Reduction --
3.3.Feature Modelling for Queries --
3.3.1.Algebra Features --
3.3.2.BGP Features --
3.3.3.Hybrid Features --
3.4.Predicting Query Performance --
3.4.1.Predictive Models --
3.4.2.Two-Step Prediction --
3.5.Experimental Evaluation and Discussion --
3.5.1.Setup --
3.5.2.Prediction Models Comparison --
3.5.3.Feature Modelling Comparison --
3.5.4.Comparison of Different Weighting Schemes in k-NN Regression --
3.5.5.Performance of Two-Step Prediction --
3.5.6.Comparison to State-of-the-Art --
3.6.Discussions Note continued: 3.7.Related Work --
3.7.1.Query Performance Prediction via Machine Learning Algorithms --
3.7.2.SPARQL Query Optimization --
3.8.Summary --
4.An Efficient Knowledge Clustering Algorithm --
4.1.Overview of Clustering with Non-negative Matrix Factorization --
4.2.Orthogonal Non-negative Matrix Factorization Over Stiefel Manifold --
4.2.1.Notations --
4.2.2.Optimization on Stiefel Manifold --
4.2.3.Update U via NRCG --
4.2.4.Update V --
4.2.5.Convergence Analysis --
4.3.Experimental Evaluation --
4.3.1.Implementation Details --
4.3.2.Data Sets --
4.3.3.Metrics --
4.3.4.Results --
4.4.Related Works --
4.5.Summary --
5.Knowledge Extraction from Unstructured Data on the Web --
5.1.Design Overview --
5.2.Source Code Topics Extraction via Topic Model and Words Embedding --
5.2.1.Data Pre-processing --
5.2.2.Topic Extraction --
5.2.3.The Coherence Measurement --
5.2.4.Automated Terms Selection for Topic Extraction --
5.3.Experimental Evaluation --
5.3.1.Setup Note continued: 5.3.2.Results --
5.4.Related Works --
5.5.Summary --
6.Building Knowledge Bases from Unstructured Data on the Web --
6.1.Design Overview --
6.2.Prototype of Knowledge Extraction from Programming Question Answering Communities --
6.2.1.Question Extraction --
6.2.2.Answer and Tags Extraction --
6.2.3.Triple Generation --
6.3.Detecting Duplicate Posts in Programming QA Communities --
6.3.1.Pre-processing --
6.3.2.Feature Modelling --
6.3.3.Binary Classification --
6.4.Experimental Evaluation and Discussions --
6.4.1.Setup --
6.4.2.Results --
6.4.3.Discussions --
6.5.Related Work --
6.5.1.Question Retrieval from QA Communities --
6.5.2.Mining PCQA Websites --
6.6.Summary --
7.Conclusion --
7.1.Summary --
7.2.Future Directions.