Web Recommendations Systems

ISBN-13: 9789811525155 / Angielski / Miękka / 2021 / 164 str.

K. R. Venugopal; K. C. Srikantaiah; Sejal Santosh Nimbhorkar

Spis treści

1 Introduction 

1.1 World Wide Web 

1.2 Web Mining 

1.2.1 Issues in Web Mining 

1.3 Web Recommendations 

1.4 Classification of Recommender system 

1.4.1 Query Recommendations 

1.4.2 Webpage Recommendations 

1.4.3 Image Recommendations 

References 

2 Web Data Extraction and Integration System for Search Engine Result Pages 

2.1 Introduction 

2.2 Related Works 

2.3 System Architecture 

2.3.1 Problem Definition 

2.4 Mathematical Model and Algorithms 

2.4.1 Web Data Extraction using Similarity Function(WDES) 

2.4.2 Web Data Integration using Cosine Similarity(WDICS) 

2.5 Experiments 

2.5.1 Precision and Recall Vs. Attributes 

2.6 Summary 

References 

3 Mining and Analysis ofWeb Sequential Patterns 

3.1 Introduction 

3.2 Related Works 

3.3 System Architecture 

3.3.1 Problem Definition 

3.4 BGCAP Algorithm 

3.5 Experiments 

3.5.1 Data Size Vs. Run Time 

3.5.2 Threshold Vs. Run Time 

3.5.3 Threshold Vs. Number of Patterns 

3.6 Summary 

References 

4 Web Caching and Prefetching 

4.1 Introduction 

4.2 Related Works 

4.3 System Architecture 

4.3.1 Problem Definition 

4.3.2 Basic Definitions 

4.4 Mathematical Model 

4.4.1 Finding Prefetching Rules using Periodicity 

4.4.2 Profit Function 

4.4.3 WCP-CMA Algorithm

4.4.4 Example

4.5 Experiments 

4.5.1 Cache Hit Ratio 

4.5.2 Delay 

4.5.3 Effect of Periodicity 

4.5.4 Effect of Cyclic Behaviour 

4.5.5 Execution Time 

4.6 Summary 

References 

5 Discovery of Synonyms from the Web

 

5.1 Introduction 

5.2 Related Works 

5.3 System Architecture 

5.3.1 Problem Definition 

5.4 System Model and Algorithm 

5.4.1 Generation of Candidate Synonyms 

5.4.2 Ranking of Candidate Synonyms 

5.4.3 ASWAT Algorithm

5.5 Experiments 

5.6 Summary 

References 

6 Ranking Search Engine Result Pages of a Specialty Search Engine

6.1 Introduction 

6.2 Related Works 

6.3 System Architecture 

6.3.1 Problem Definition 

6.4 Mathematical Model 

6.4.1 Probability of Correctness of Facts (PCF) 

6.4.2 Implication Between Facts 

6.4.3 SIM (TF,F0 ) for Books Domain 

6.5 Complexity Analysis 

6.5.1 Time Complexity 

6.5.2 Space Complexity 

6.6 Experiments 

6.7 Summary 

References 

7 Construction of Topic Directories 

7.1 Introduction 

7.2 Related Works 

7.3 System Architecture 

7.3.1 Problem Definition 

7.4 Mathematical Model and Algorithm 

7.4.1 Hashing 

7.4.2 Levenshtein Distance (LD) 

7.4.3 Levenshtein Similarity Weight (LSW) 

7.4.4 Similarity betweenWebpage and Category in Web Directory 

7.4.5 Mapping of Pages onto Categories 

7.4.6 Algorithm MPCLSW

7.5 Experiments 

7.5.1 Execution Time 

7.5.2 Accuracy 

7.5.3 Precision 

7.5.4 Recall 

7.5.5 F-score 

7.6 Summary 

References 

8 Query Relevance Graph for Query Recommendations 

8.1 Introduction 

8.2 Related Works 

8.2.1 Query Expansion 

8.2.2 Snippet based Query Recommendations 

8.2.3 Graph based Query Recommendations 

8.2.4 Recommendation Applications 

8.3 Query Relevance Model and QRGQR Algorithm 

8.3.1 Problem Definition 

8.3.2 Query Click Graph 

8.3.3 Query Text Similarity Graph 

8.3.4 Query Relevance Graph 

8.3.5 QRGQR Algorithm 

8.4 Experiments 

8.4.1 Data Collection 

8.4.2 Data Cleaning 

8.4.3 Varying of Parameter-Jaccard Coefficient 

8.4.4 Query Recommendation results 

8.4.5 Performance Analysis 

8.4.6 Efficiency 

8.4.7 Image Recommendation 

8.5 Summary 

References 

9 Related Search Recommendation with User Feedback Session 

9.1 Introduction 

9.2 Related Works 

9.2.1 Measuring Similarity between Two Words 

9.2.2 Query Recommendation Techniques 

9.3 Related Search Recommendation Framework and RSR Algorithm

9.3.1 Problem Definition 

9.3.2 Co-occurrenceMeasures to Compute Semantic Similarity

9.3.3 WordNet based Semantic Similarity

9.3.4 Rocchio’s Model 

9.3.5 Snippet Click Model 

9.3.6 RSR Algorithm 

9.4 Experiments 

9.4.1 Data Collection 

9.4.2 Experiment Setup 

9.4.3 Query Recommendation Results 

9.4.4 Performance Analysis 

9.5 Summary 

References

10 Webpage Recommendations based Web Navigation Prediction 

10.1 Introduction 

10.2 Related Works 

10.2.1 Web Page Prediction 

10.2.2 Prediction Applications 

10.3 Web Navigation Prediction Framework and WNPWR Algorithm 

10.3.1 Problem Definition 

10.3.2 Session Identification Method with Average Time of Visiting Web Pages 

10.3.3 Prediction Models 

10.3.4 Two-Tier Prediction Framework 

10.3.5 WNPWR Algorithm

10.4 Experiments 

10.4.1 Data Collection 

10.4.2 User and Session Identification 

10.4.3 Experiment Setup 

10.4.4 Results Comparison 

10.5 Summary 

References 

11 Webpage Recommendations based on User Session Graph 

11.1 Introduction 

11.2 Related Works 

11.3 Webpage Recommendations Framework andWRUSG Algorithm

11.3.1 Problem Definition 

11.3.2 Webpage Recommendations Framework 

11.3.3 WRUSG Algorithm 

11.4 Experiments 

11.4.1 Data Collection 

11.4.2 Experiment Set-up 

11.4.3 PerformanceMetrics 

11.4.4 Performance Evaluation 

11.5 Summary 

References 

12 Advertisement Recommendations using ExpectationMaximization

12.1 Introduction 

12.2 Related Works 

12.3 Prediction Conversion in Advertising using Expectation Maximization Model and PCAEM Algorithm 

12.3.1 Problem Definition 

12.3.2 Prediction Conversion in Advertising using Expectation Maximization Model 

12.3.3 PCAEM Algorithm 

12.4 Experiments 

12.4.1 Data Collection 

12.4.2 Experiment Setup 

12.4.3 PerformanceMetrics 

12.4.4 Performance Evaluation 

12.5 Summary 

References 

13 Image Recommendations with Absorbing Markov Chain 

13.1 Introduction 

13.2 Related Works 

13.2.1 Content Based Image Retrieval 

13.2.2 Annotation Based Image Retrieval 

13.3 Image Recommendation Framework and IRAbMC Algorithm 

13.3.1 Problem Definition 

13.3.2 Image Recommendation Framework 

13.3.3 IRAbMC Algorithm 

13.4 Experiments 

13.4.1 Data Collection 

13.4.2 Experiment Setup 

13.4.3 Performance Evaluation 

13.5 Summary 

References 

14 Image Recommendation with User Relevance Feedback Session 

14.1 Introduction 

14.2 Related Works 

14.3 Image Recommendation Framework and IR URFS VF Algorithm

14.3.1 Problem Definition 

14.3.2 Image Recommendation Framework 

14.3.3 IR URFS VF Algorithm 

14.4 Experiments

14.4.1 Data Collection 

14.4.2 Experiment Setup 

14.4.3 Performance Evaluation 

14.5 Summary 

References 

15 Image Recommendation by ANOVA Cosine Similarity 

15.1 Introduction 

15.2 Related Works 

15.2.1 Content Based Image Retrieval (CBIR) 

15.2.2 Annotation Based Image Retrieval (ABIR) 

15.2.3 Text + Visual (Hybrid) method for image search 

15.2.4 Image Search with reduced Semantic Gap 

15.3 ACSIR Framework and Algorithm 

15.3.1 Problem Definition 

15.3.2 ACSIR Framework 

15.3.3 ACSIR Algorithm

15.4 Experiments 

15.4.1 Data Collection 

15.4.2 Experiment Set-up 

15.4.3 Performance Evaluation 

15.5 Summary 

References 

Index 

O autorze

Dr. K R Venugopal is the Vice Chancellor of Bangalore University. He holds eleven degrees, including a Ph.D. in Computer Science Engineering from IIT-Madras, Chennai and a Ph.D. in Economics from Bangalore University. He also has degrees in Law, Mass Communication, Electronics, Economics, Business Finance, Computer Science, Public Relations and Industrial Relations. He has authored and edited 68 books and published more than 800 papers in refereed international journals and international conferences. Dr. Venugopal was a postdoctoral research scholar at the University of Southern California, USA. He has been conferred with IEEE fellow and ACM Distinguished Educator for his contributions to computer science engineering and electrical engineering education.

Dr. K C Srikantaiah is a Professor at the Department of Computer Science and Engineering at SJB Institute of Technology, Bangalore, India. He received his B.E. from Bangalore Institute of Technology, M.E. from University Visvesvaraya College of Engineering, Bangalore, in 2002 and Ph.D. degree in Computer Science and Engineering from Bangalore University in 2014. He has published 20 research papers and authored a book on Web mining algorithms. His research interests include data mining, Web mining, big data analytics, cloud analytics and the Semantic Web.

Dr. Sejal Santosh Nimbhorkar is an Associate Professor at B N M Institute of Technology. She has more than 15 years of industry, research and teaching experience. She holds M.E. and B.E. degrees in Computer Science and Engineering from University Visvesvaraya College of Engineering and Gujarat University, respectively. She has published 18 papers in refereed international journals and international conferences. She received an outstanding paper award at the 2015 European Conference on Data Mining. Dr. Nimbhorkar has also received project grants from Karnataka State Council for Science and Technology (KSCST). Her research interests include mining, Web mining, sentiment analysis and IoT.

Blurb

This book focuses on Web recommender systems, offering an overview of approaches to develop these state-of-the-art systems. It also presents algorithmic approaches in the field of Web recommendations by extracting knowledge from Web logs, Web page content and hyperlinks. Recommender systems have been used in diverse applications, including query log mining, social networking, news recommendations and computational advertising, and with the explosive growth of Web content, Web recommendations have become a critical aspect of all search engines.

 

The book discusses how to measure the effectiveness of recommender systems, illustrating the methods with practical case studies. It strikes a balance between fundamental concepts and state-of-the-art technologies, providing readers with valuable insights into Web recommender systems.