Literary Analysis and Word Cloud, Topic Modeling Natural Language Processing Applications

Word Cloud and Topic Modeling

Word Cloud

• A tool for analyzing words obtained from data.
• A text mining technique for analyzing word frequency and main keywords.

Topic Modeling

• One of the statistical models for analyzing text based on words that appear frequently on the surface.
• A tool for extracting topics contained in a corpus of sentences.
• One of the statistical models for discovering abstract topics of a corpus called topics.
• A text mining technique used to discover the hidden semantic structure of text.

Text Mining

• A process of deriving new information from text from a new angle, as a method of extracting and analyzing useful information from text data.

Analysis of Literary Works from a New Angle

Traditional literary analysis relies on the researcher's extensive reading experience and literary analysis to infer and define the theme based on the main character and events (Lee Jae-yeon, 2016: 282).

Text mining, on the other hand, is:

• A theoretical attempt to find the hidden essence that the text "represents" regardless of the author's intention.
• Methodological development (Jeong Seong-hun, 2021: 112) by statistical algorithm.
• Identifying topics based on the similarity between keywords objectively extracted by the computer.

Word Cloud and Literary Work Analysis

Methodology

• "Writers don’t waste a single word or punctuation mark." - Lee Kang-ryong (Professor, Department of Korean Language and Literature, Sogang University)
• Why did the author use so many of these words?
• What do these words, which the author unconsciously or intentionally uses a lot, mean?

Process

1. Data collection: Collect text that can be read by a computer in a form that can be processed.
2. Data preprocessing:
   • Unnecessary words are analyzed to prevent them from skewing the analysis.
   • This involves lowercase conversion, stop words (stopwords) removal, special character removal, morphological analysis and extraction of titles.
   • Typically handled using Python or R.
   • For printed copies, OCR is used to convert them to a computer-readable txt file.
3. Create a word cloud: Using preprocessed data, a word cloud is generated, with frequently occurring words appearing more prominently.
4. Results analysis: Analyzing literary works from a new angle, centered on the word cloud.

Word Cloud and Literary Work Analysis: Example - One Hundred Years of Solitude

Spatial meaning in the word 'CASA'

• Related to magical realism.
• The space where most events take place, including the birth and death of the Buendia family.
• Both ordinary and strange; an everyday space for sleeping, eating, resting, and making a living.
• Functions as a device that makes readers accept the bizarre reality gently.
• Examples:
  • The ghost of Prudencio Aguilar walks through the real world.
  • The dead Melquiades returns from the dead and constantly communicates with living people, including Aureliano.

Spatial meaning in the word 'Cuarto'

• Regarding loneliness in Latin America
• Meaning a very personal and private space within the home.
• An isolated space that neither influences nor is influenced by the outside world.
• Examples:
  • Jose Arcadio Buendia is absorbed in alchemy.
  • Colonel Buendia is hiding to avoid external attacks.
  • A place where death comes to an end.
  • Aureliano's parchment left behind by Melquiades where to decipher and meet the ending.

The Solitude of the Buendia Family \longrightarrow The Loneliness of Latin Americans

Temporal meaning in the word 'Noche'

• Criticism of Westernization and Power
• Dark, quiet and secret time
• There are incidents that are secretive and have negative nuances.
• Examples:
  • The time when Jose Arcadio leaves with the gypsies who come to the village at night.
  • The time when the sole survivor of the protest returns.
  • The time when police and government forces stormed the house and shot Buendia dead, while the police arrested the ringleaders, including Jose Arcadio Segundo.
  • Colonel Aureliano Buendia's time in the war, when he carried out many surprise attacks.
  • A depiction of power secretly committing injustice at night while everyone is asleep.

Western powers that invaded Latin America and suppressed the existing culture and people, and secretly killed people. A time to symbolically reveal the oppressive state power

The eventual meaning of the word 'Guerra'

• Reflecting on Latin America's modern and contemporary history
• The war against the government forces led by Colonel Aureliano Buendia
• The liberals unilaterally oppose the conservatives.
• All comrades are absorbed or eliminated by the government.
• Fighting between rebels and government forces
• Struggle
• Unfinished war, unfinished business
• They fought against the existing powers but gave up the war on their own and surrendered to the conservatives, bringing about no change.
• Even people in Macondo who had nothing to do with the war became involved.
• Reflecting the times in Latin America

Conclusion

Words that symbolically express the theme of One Hundred Years of Solitude

Topic Modeling and Literary Work Analysis

Methodology

1. Data collection: Collect text that can be read by a computer in a form that can be processed.
2. Data preprocessing:
   • Unnecessary words are analyzed to prevent them from skewing the analysis.
   • This involves lowercase conversion, stop words (stopwords) removal, special character removal, morphological analysis and extraction of titles.
   • The number of topics that will appear at this time is set at the analyst's discretion.
   • Therefore, background knowledge of the Korean analyst is required in the text to be analyzed.
   • Typically handled using Python or R.
   • For printed copies, OCR is used to convert them to a computer-readable txt file.
3. Create topic modeling: Using preprocessed data, create a topic model to analyze the topics being discussed.
4. Results analysis: Analyze the words that appear in each topic and interpret its meaning.

Topic Modeling and Literary Work Analysis: Example - Tim O'Brien's 'The Things They Carry'

• A realistic novel that tells the story of the tragedy of the Vietnam War.
• A collection of short stories by Vietnam War veteran writer Tim O'Brien.
• It exposes the many facets of the Vietnam War, which was full of contradictions and dilemmas.
• It contains 22 short stories that depict the extreme situation of war.
• The author crosses the line between fact and fiction, and It pierces through iron.
• Also, the comical gestures, desires, and loneliness of American soldiers bursting out in the most desperate situation of war.

Topic Modeling Results

The ten topics can be broadly classified into three themes.

1. Characteristics of the Vietnam War: Topics 1, 2, 4, 5, 8, and 9 define the characteristics of the Vietnam War.
   • The war is not a noble war that establishes the value of exceptionalism but a war that is mired in a swamp filled with rain, 'shit', and 'muck', implying a situation in which the United States is unable to get out of it.
2. Regaining Identity: Topics related to how soldiers must regain their identity, purity, and loved ones that they lost in Vietnam.
   • Topic 3: "The life of the dead" is related to the protagonist's first love, Linda.
   • Topic 9: "Songtra Peak's Lover" is connected to Mary Anne.
   • Linda and Mary Anne symbolize the lost comrades of the American soldiers who fought in Vietnam, and their own purity and humanity, and through this, the soldiers are thinking about how to recover what they have lost.
3. Emotions and Situations: Topic 4 contains all the important topics (story, war, smell, truth, rice, Vietnam).
   • It shows the emotions and situations felt by the readers through Kathleen, the protagonist's daughter, who symbolizes the later generation that does not understand the Vietnam War.