Please use this identifier to cite or link to this item:
https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4463
Title: | Estimating Task Complexity of Text Analysis Tasks |
Authors: | Chathurika, W.M.T. |
Issue Date: | 5-Aug-2021 |
Abstract: | Text analysis is one of the most common approaches in machine learning applications. The process of analyzing raw data to make conclusions on those data and to find trends and answers for some questions which are known as data analytics captures a broad scope in the field of computing. “Text Analysis” is the term that is used to describe the process of computational analysis of text data. It involves numerous techniques and approaches to bring text data to an end where they can be mined for trends, patterns, or insights. The accuracy of machine learning algorithms depends on the size of the train data set. The quantity of data is affected by various factors. It depends on the complexity of the problem, training method, and diversity of inputs. Depending on the type of data, they can be expensive. Due to that, it is useful to know the amount of data needed before training a model. In this paper, the broad area of text analytics would be broadened down to text classification to reduce the complexity in the experimental approach. In this research text classification algorithms will be trained using different datasets, that consist of different amounts of data and different features to observe the accuracies. |
URI: | http://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4463 |
Appears in Collections: | 2020 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
2017 MCS 014.pdf | 2.63 MB | Adobe PDF | View/Open |
Items in UCSC Digital Library are protected by copyright, with all rights reserved, unless otherwise indicated.