Development of ML‐Based Methodologies for Adaptive Intelligent E‐Learning Systems and Time Series Analysis Techniques
Indra Kumari, Indranath Chatterjee, Minho Lee · 2023
The use of machine learning (ML) and artificial intelligence (AI) to solve problems is rapidly replacing traditional methods in many academic and business sectors (DL). Learning techniques such as search, logic, and probability are combined with more modern methods such as (deep), (un) supervised, and reinforcement learning in computational models. These works provide effective strategies for incorporating machine learning (ML) models into an e-learning system to match the most relevant e-content with the interests of individual students. In the fields of ML, AI, data mining, and pattern recognition, the evaluation and selection of learning methods is a hotspot for investigation. There is a plethora of algorithms for learning available in the literature on ML, and more are being added every day. However, choosing the optimal learning algorithm for a certain dataset is challenging. There is a requirement for an adaptable learning system due to the proliferation of available learning algorithms and the dynamic nature of data sets. The sources of time series data are ubiquitous in modern life. Biological signals, weather recordings, stock exchange rates, and many other types of information are continuously produced by a vast array of human activities and practical applications. Time series information may be either univariate or multivariate. Time series analysis gave birth to several new fields of study, such as indexing, predicting, categorizing, and grouping. In this chapter, we examine the problems of representing, classifying, and predicting time series.