Automated Assessment of Lymphocytes Using Machine Learning Techniques

EasyChair Preprint 5329

Abstract

This paper investigates the performance of machine learning techniques for automated assessment of lymphocytes. A total of four algorithms, i.e., support vector machine (SVM), deep learning for java (DL4J), multi-layer perceptron (MLP) and K* are applied to the lymphocytes dataset. To ensure the robustness, all the four algorithms are evaluated with both pre-processed data and data without prior pre-processing. With the pre-processed data, MLP outperforms all the other techniques and achieve an accuracy of 98.64%. SVM, DL4J and K* achieve accuracies of 97.97%, 96.62% and 97.29%, respectively.

Keyphrases: DL4J, K*, Lymphocytes, MLP, SVM, machine learning

@booklet{EasyChair:5329,
  author    = {Afaq Ahmad and Azmat Ullah and Kaleem Nawaz Khan and Muhammad Salman Khan},
  title     = {Automated Assessment of Lymphocytes Using Machine Learning Techniques},
  howpublished = {EasyChair Preprint 5329},
  year      = {EasyChair, 2021}}