DATA CLASSIFICATION USING VARIOUS LEARNING ALGORITHMS
Need help with a related project topic or New topic? Send Us Your Topic
DOWNLOAD THE COMPLETE PROJECT MATERIAL
DATA CLASSIFICATION USING VARIOUS LEARNING ALGORITHMS
Chapter One: Introduction 1.1 Background of the Study
The rapid increase in data volume and variety makes it difficult to extract usable information from massive datasets. Extracting relevant information and hidden patterns from data can be tricky, but increasingly vital [1].
Research in fields such as biology, astronomy, engineering, consumer transactions, and agriculture involves daily monitoring of large datasets.
Large datasets pose issues for traditional statistical analysis techniques. The most challenging aspect is managing the multiple variables (dimensions) associated with each observation. Reducing the number of dimensions in high-dimensional datasets can enhance analytical accuracy and efficiency [2].
It is beneficial to be able to map a set of points (n) in d-dimensional space onto a p-dimensional space (where p << d) without significantly altering the points’ inherent attributes, such as inter-point distances and labels. This is known as dimensionality reduction [3].
There are numerous approaches for lowering the dimensionality of data. These methods fall into two kinds. In the first, the reduced dataset’s attributes are linear combinations of the original dataset’s attributes.
In the second category, the reduced dataset contains only a subset of the original dataset’s attributes [4]. There are two categories of techniques: Random Projection (RP), Singular Value Decomposition (SVD), and Principal Component Analysis (PCA).
The second category includes the Combined Approach (CA), Direct Approach (DA), Variance Approach (Var), New Top-Down Approach (NTDn), New Bottom-Up Approach (NBUp), New Top-Down Approach (modified version), and New Bottom-Up Approach (modified version) [5].
Dimensionality reduction reduces high-dimensional data to a compact representation that retains only essential information, making it suitable for machine learning algorithms that struggle with large datasets [6].
Calculating inter-point distances is crucial for machine learning tasks. However, as dimensionality increases, the distance between a sample point and the nearest point becomes similar to the distance between the sample point and the most distant point, negatively impacting the performance of machine learning algorithms [7].
Dimensionality reduction is a crucial preprocessing step for many machine learning methods.
Machine learning refers to the ability of computer systems to learn and improve their computations via experience [8, 9]. There are two basic types of machine learning algorithms: supervised and unsupervised.
These algorithms have been utilised to solve complicated real-world issues [10, 11]. Unsupervised learning categorises observations into clusters based on their similarities. This categorization is also known as clustering [8]. K-means clustering is well-known for its ability to handle huge datasets [12].
Classification, unlike clustering, is a supervised learning method that predicts the label for any valid input based on a “training set” of instances [8], [12]. This inquiry examines two types of classification algorithms: eager and lazy learners. Eager learning algorithms aim to 3
Need help with a related project topic or New topic? Send Us Your Topic