Robust Latent Feature Learning for Incomplete Big Data

Latent feature analysis (LFA) is a popular representation learning method for analyzing incomplete big data due to its high accuracy and scalability. This book introduces robust latent feature learning methods to address uncertainty caused by incomplete characteristics,such as robust latent feature learning based on smooth L1-norm,improving robustness using L1-norm,data-characteristic-aware latent feature learning,posterior-neighborhood-regularized latent feature learning,and generalized deep latent feature learning. It also provides algorithms and real application cases to help students, researchers, and professionals build robust models to analyze incomplete big data.

Format: Paperback / softback
Length: 112 pages
Publication date: 08 December 2022
Publisher: Springer Verlag, Singapore

Incomplete big data is a common occurrence in numerous industrial applications, including recommender systems, the Internet of Things, intelligent transportation, cloud computing, and more. Analyzing these data for extracting valuable knowledge and patterns is of paramount importance. Latent feature analysis (LFA) stands out as a widely adopted representation learning method for handling incomplete big data due to its remarkable accuracy, computational efficiency, and scalability. However, existing LFA methods fail to fully account for the uncertainty inherent in incomplete data.

In this comprehensive book, the author presents a series of robust latent feature learning methods designed to tackle this uncertainty effectively and efficiently. These methods encompass robust latent feature learning based on smooth L1-norm, improving robustness of latent feature learning using L1-norm, enhancing robustness of latent feature learning using double-space, data-characteristic-aware latent feature learning, posterior-neighborhood-regularized latent feature learning, and generalized deep latent feature learning. By delving into these approaches, readers gain a comprehensive understanding of the challenges associated with analyzing incomplete big data and the strategies employed to build robust models for tackling such data.

Furthermore, the book offers a wealth of algorithms and real-world application cases, making it an invaluable resource for students, researchers, and professionals seeking to build their models for analyzing incomplete big data. With its clear explanations, practical examples, and comprehensive coverage, this book empowers individuals to harness the power of latent feature learning and make meaningful insights from incomplete data.

Weight: 209g
Dimension: 235 x 155 (mm)
ISBN-13: 9789811981395
Edition number: 1st ed. 2023