Sentiment Analysis of Code-Mixed Text: A Review

Abstract

In recent times, sentiment analysis has become one of the most active research and progressively popular areas in information retrieval and text mining. To date, sentiment analysis has been applied in various domains such as product, movie, sport and political reviews. Most of the previous work in this field has focused on analyzing only a single language, especially English. However, with the need of globalization and the increasing number of the Internet used worldwide; it is common to see the post written in multiple languages. Moreover, in an unstructured content like Twitter posts, people tend to mix languages in one sentence, which make sentiment analysis process even harder and more challenging. This paper reviews the state-of-the-art of sentiment analysis for code-mixed, which includes the detail discussions of each focus area, qualitative comparison and limitations of current approaches. This paper also highlights challenges along this line of research and suggests several recommendations for future works that should be explored.

Sentiment Analysis in Low-Resource Settings: A Comprehensive Review of Approaches, Languages, and Data Sources

“…The scarcity of language tools, such as part-of-speech (POS) taggers, and NLTK, tailored for low-resource languages poses a significant constraint, limiting the exploration of diverse sentiment analysis methods [130]. Furthermore, the prevalent phenomenon of code-mixing or code-switching in multilingual low-resource communities and social media interactions [30,65,131], has witnessed a notable upswing over the past decade [132]. However, these individuals who belong to multiple online communities demonstrated a flexible ability to adjust and switch their patterns of responding to compliments based on the specific online cultural context in which they were engaged [133].…”

“…However, these individuals who belong to multiple online communities demonstrated a flexible ability to adjust and switch their patterns of responding to compliments based on the specific online cultural context in which they were engaged [133]. This practice involves the integration of languages with differing linguistic resources within the same textual content [30], resulting in complex linguistic features such as intra-sentential and inter-sentential code-mixing, inventive spelling variations, lexical borrowing, and phonetic typing [65]. These limitations underscore the critical necessity for comprehensive research endeavours in the arena of low-resource languages and sentiment analysis, given their escalating influence in the digital landscape.…”

“…Sentiment classification methods are the methods used in categorizing text into polarities. They are classified into three categories: lexicon-based, machine learning (ML), and hybrid [12,24,[28][29][30][31][32][33][34][35]. However, the lexicon-based method relies on a predefined set of patterns, often referred to as a sentiment dictionary or lexicon, where each data entry is linked with a specific sentiment orientation [17,36].…”

Comparative Analysis of Transformer Models for Sentiment Analysis in Low-Resource Languages

“…Sentiment analysis classification has different techniques, which are classified into three categories or classes lexiconbased, machine learning-based, and hybrid-based [9], [25], [26], [27], [28]. The machine learning techniques leverage well-known ML algorithms or models to address sentiment analysis.…”

Language Threshold for Multilingual Sentiment Analysis System

“…However, the success of this approach depends heavily on the language model used. Creating a sentiment lexicon or dictionary [8] is time-consuming, as it requires human annotation.…”