Sayed-Mouchaweh M. (ed.) Learning from Data Streams in Evolving Environments. Methods and Applications

Springer, 2019, — 320 p.The volume of data is rapidly increasing due to the development of the technology of information and communication. This data comes mostly in the form of streams. Learning from this ever-growing amount of data requires flexible learning models that self-adapt over time. In addition, these models must take into account many constraints: (pseudo) real-time processing, high-velocity, and dynamic multi-form change such as concept drift and novelty. Consequently, learning from streams of evolving and unbounded data requires developing new algorithms and methods able to learn under the following constraints: (1) random access to observations is not feasible or it has high costs, (2) memory is small with respect to the size of data, (3) data distribution or phenomena generating the data may evolve over time, which is known as concept drift and (4) the number of classes may evolve overtime, which is known as concept evolution. Therefore, efficient data streams processing requires particular drivers and learning techniques able to perform:
Incremental learning in order to integrate the information carried out by each new arriving data;
Decremental learning in order to forget or unlearn the data samples which are no more useful;
Novelty detection in order to learn new concepts.
This book presents and discusses recent advanced techniques, methods and tools treating the problem of learning from data streams generated by evolving and non-stationary phenomena. These methods address the different challenges (with concept drift, with concept evolution, with both concept drift and concept evolution) of learning from multidimensional data streams using classification or clustering techniques, mono model or ensemble, online or semionline, centralized processing or distributed computing, instance based or window based, and incremental decremental with or without transfer learning for different applications (social networks, Twitter data analysis, stream trends and dynamics visualization, user query and preference evaluation, gene network, customer relationship management, electricity price prediction, taxi traffic management, etc.).Transfer Learning in Non-stationary Environments
A New Combination of Diversity Techniques in Ensemble Classifiers for Handling Complex Concept Drift
Analyzing and Clustering Pareto-Optimal Objects in Data Streams
Error-Bounded Approximation of Data Stream: Methods and Theories
Ensemble Dynamics in Non-stationary Data Stream Classification
Processing Evolving Social Networks for Change Detection Based on Centrality Measures
Large-Scale Learning from Data Streams with Apache SAMOA
Process Mining for Analyzing Customer Relationship Management Systems: A Case Study
Detecting Smooth Cluster Changes in Evolving Graph Structures
Efficient Estimation of Dynamic Density Functions with Applications in Data Streams
Incremental SVM Learning: Review
On Social Network-Based Algorithms for Data Stream Clustering