PHD Research Proposal in Machine Learning

The machine learning techniques have prompted to the forefront over the last few years due to the advent of big data [1]. Machine learning is a precise subfield of artificial intelligence (AI) that seeks to analyze the massive data chunks and facilitate the system to learn the data automatically without the explicit support of the programming. The machine learning algorithms attempt to reveal the fine-grained patterns from the unprecedented data under the multiple perspectives and to build an accurate prediction model as never before [2]. By the purpose of learning, the machine learning algorithm is sub-categorized into four broad groups include supervised learning, semi-supervised learning, unsupervised learning, and reinforcement learning [3]. Whenever the new unseen data is fed as input to the machine learning algorithm, it automatically learns and predicts the forthcoming by exploiting the previous experience over time. Machine learning is continually liberating its potency in a broad range of applications including Internet of Things (IoT), computer vision, natural language processing, speech processing, online recommendation system, cyber security, neuroscience, prediction analytics, fraud detection, and so on [4].

The machine learning algorithm is capable of managing multi-dimensional data under the dynamic environment. Despite its advantages, there are still difficulties and challenges. The machine learning algorithm still requires an additional mechanism for predicting the massive number of new classes. Also, the developed analytical abilities given by the machine learning algorithm pose new difficulties and challenges while managing privacy. The lack of adequately annotated raw data complicates the learning process and results in inaccurate results. Furthermore, the reliable use of a machine learning algorithm relies on human experts. Interpretation of outcomes is a significant challenge in the machine learning algorithm.
The machine learning technique enables the hidden learning insights at multi-granularity of big data. Even though, It suffers from the issue of high dimensionality, distributed computing, scalability, adaptability, and the streaming data. In addition, the duplication of data also creates a major impact on the learning of the machine learning algorithm. In many cases, the multi-class classification for the evolving large-scale data induces the complexity. The main issue of the machine learning algorithm is its vulnerability to errors. Furthermore, machine learning techniques are lacks in variability. The machine learning algorithm entails accurate details of experience to predict the future. Moreover, it requires large-scale data to learn the disparate topics that consume much time and resources.

Reference

  • [1] Qiu, Junfei, Qihui Wu, Guoru Ding, Yuhua Xu, and Shuo Feng, “A survey of machine learning for big data processing”, EURASIP Journal on Advances in Signal Processing 2016, No.1, pp.67, 2016.

  • [2] Royal Society (Great Britain), ”Machine Learning: The Power and Promise of Computers that Learn by Example: an Introduction” Royal Society, 2017..

  • [3] Ayodele, Taiwo Oladipupo, “Types of machine learning algorithms” In New advances in machine learning, IntechOpen, 2010.

  • [4] Das, Kajaree, and Rabi Narayan Behera, “A survey on machine learning: concept, algorithms and applications”, International Journal of Innovative Research in Computer and Communication Engineering, Vol5, No.2, pp.1301-1309, 2017.

Leave Comment

Your email address will not be published. Required fields are marked *

clear formSubmit