Over the past two decades, literary works from Afghanistan have sought to depict the extensive and increasing level of violence directed against individuals and communities. This study aims to identify the different types of violence represented in selected literary works from the country. To this end, an artificial intelligence approach—comprised of opinion mining and hybrid machine learning models—is developed as a method of analysis. The Afghan American writer Khaled Hosseini's novels—The Kite Runner (2003), A Thousand Splendid Suns (2007), and And the Mountains Echoed (2017)—are used as primary data sources. As a theoretical framework, the study benefits from the Norwegian sociologist Johan Galtung's theory of conflict triangle—direct violence, structural violence, and cultural violence. To classify the sentiments related to types of violence in the selected novels, the study utilizes sentiment analysis and machine learning algorithms.
TopIntroduction
Over the past two decades, literary works from Afghanistan have sought to depict the extensive and increasing level of violence directed against individuals and communities. Since the invasion of the country by US-led forces in late 2001, Afghanistan has become a locus of uncertain political and religious tensions, fear, conflict, and violence. The high levels of violence in the country are often portrayed by literary writers through fictional narratives (Asl, 2019). The novelist Khaled Hosseini is among the forerunners of Afghan writers whose works are replete with instances of violence prevalent in the country. His three novels The Kite Runner (2003), A Thousand Splendid Suns (2007), and And the Mountains Echoed (2017) provide illuminating insights about the daily experiences of conflict and violence in contemporary Afghanistan. This study aims to analyze the three novels to identify the different types of violence represented in the stories in light of the Norwegian sociologist Johan Galtung’s theory of Conflict Triangle—direct violence, structural violence, and cultural violence.
To achieve this goal, an artificial intelligence approach—comprised of Opinion Mining and Hybrid Machine Learning Models—is developed as a method of data mining and data analysis. More specifically, the study utilizes Opinion Mining, Sentiment Analysis, and Machine Learning algorithms to identify and classify the sentiments related to the various types of violence as depicted in the selected novels. Opinion Mining or Sentiment Analysis are useful in analyzing NLP applications. With the development of applications like network public opinion analysis, the demand for sentiment analysis and opinion mining has been growing. OM is one of the most interesting applications of Natural Language Processing (NLP) used in social media applications whose goal is the evaluation and classification of text with emotionally charged language which expresses or implies positive, negative, and neutral sentiments. Sentiment analysis in a broad sense is a way to know a user's emotion (sentiment) using his features (Oramas Bustillos et al., 2019). Positive emotion can be happy or liking something while negative emotions include anger, unhappiness, etc. Neutral emotions are when a person shows no expression (Asri et al., 2022a; Fasha et al., 2022a; Sofian et al., 2022a; Suhendra et al., 2022a)
The main objective of this chapter is to identify the different types of violence in selected literary books based on Johan Galtung’s framework. In addition, this chapter attempt to classify the sentiments in the selected novels using sentiment analysis. Finally, this chapter aims to classify sentiments related to the types of violence as detected in objective one using machine learning.
Other sections of this chapter are structured as follows. Section 2 presents a review of literature, while Section 3 discusses the methodology of the study. Section 4 is the analysis and results, and section 5 provides the discussions. Finally, the conclusions and future works are discussed in sections 6.