Academic journal article South Asian Studies

The Text of the Memorandum 'Sikhs and the New Constitution for India' (1930). Political Importance and Linguistic Sentiment Analysis

Academic journal article South Asian Studies

The Text of the Memorandum 'Sikhs and the New Constitution for India' (1930). Political Importance and Linguistic Sentiment Analysis

Article excerpt

Abstract

Communication research has often been performed through text analysis and content analysis. They are research techniques that vary in nature and allow the researcher to make inferences from data present in a text and extract relevant information from them. Content analysis has been used in a variety of contexts with diverse research objectives, goals and methods including computerized and automated methods. In our textual analysis we have selected the Sikh Memorandum from the Indian Round Table Conference (1930) in order to extract linguistic information about its contents that can be related to its historical context through automated methods and in this manner find out word patterns to denote sentiment importance. The purpose is to find out the agreement of some of the automated online sentiment analysis tools that are available on the net as well as whether the Memorandum had positive, neutral or negative polarity. The results will be analyzed.

Keywords: Indian Round Table Conference, Punjab, Sikhs, linguistic analysis of texts, sentiment analysis, UMIGON, Python, LIWC, computational linguistics.

Introduction

Methods of communication have increased with the improvements in technology. In this sense textual analysis methods have grown in a parallel manner from a beginning of manual methods to nowadays automated methods. These methods do pursue multiple uses and linguistic goals. Qualitative content analysis is one of multiple research methods which are nowadays utilized to analyze data from texts. Some of the other methods utilized for this purpose are historical research, phenomenology, ethnography, grounded theory, etc. Content analysis focuses on the physiognomies of language as communication with concentration on the subject matter or contextual meaning of the analyzed corpus or text. (Budd, Thorp, & Donohew, 1967; Lindk vist, 1981; McTavish &Pirro, 1990; Tesch, 1990).

The purpose of content analysis is to provide knowledge and understanding of the phenomenon under study. We utilized the directed content analysis approach. (R.W, R.K, & L, 1967)

Every text presents four levels through which it can be interpreted:

* Grammatical: Cultural experience and training in the usage of symbols.

* Cognitive schemas: Speaks about the experience of the individual and its knowledge organization.

* Stylistic/means of expression: Speaks of the subject's education, cultural inheritance and creative abilities.

* -Conceptual (array of concepts with innovative meanings): Compares the effects of interpretation of the previous levels. This level of interpretation is more intellectual but also intuitive. (K, 1981)

In addition, the lexicon of any language presents level of polarity; whether the words are positive, neutral or negative. Furthermore, each sentence also presents a level of polarity. That is, whether a sentence is positive, neutral or negative. In some cases a sentence may present itself with positive and negative elements within it. Then, it is considered as both, positive and negative. It is for this reason that sentences or expressions can be considered, positive, negative, both or neutral. (K.E, 1981)

A large percentage of the work on sentiment analysis is based on whether a text is positive or negative (Turney, 2002)

On the contrary, our research utilizes online sentiment analyzing software and tests the accuracy of two major online software tools utilized for this purpose.

The software programs in question are UMIGON (Clement, 2013) and Python (NLTK).1 UMIGON was developed in 2012 and has the purpose of analyzing sentiment in tweets. Python, on the other, hand uses a hierarchical system of classification which determines whether the text is neutral first and then the level of its polarity only in the case of a non-neutral text.

In order to test the accuracy of these two software tools we will introduce sentences with the purpose of obtaining a result. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.