Results of the summarizing process for the arXiv paper: 1912.10847v1

This paper explores the similarities between various sacred texts using text mining techniques. The main source of religious teachings is their sacred texts, which can vary based on factors such as geographical location or the time of the birth of a particular religion. Despite these differences, there may be similarities in the lessons taught by these texts. The corpus used in this study consists of Asian texts (Tao Te Ching, Buddhism, Yogasutra, Upanishad) and non-Asian texts (four Bible texts). To measure similarity, different measures like Euclidean, Manhattan, Jaccard, and Cosine are applied to raw Document Term Frequency (DTM) and normalized DTM. These measures reveal similarity based on word usage. Additionally, supervised learning algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Random Forest are used to predict the correct sacred text for any given chapter in the corpus. The study also includes K-means clustering visualizations on Euclidean distances of raw DTM. These visualizations reveal a pattern of similarity among the sacred texts, with Upanishads and Tao Te Ching being the most similar texts in the corpus. The research aims to find if there are any similarities between various sacred texts in terms of what they teach and how they teach religious lessons. Text mining using machine learning and feature extraction is employed to identify patterns in document collections. The study applies similarity measures such as Euclidean, Manhattan, Jaccard, and Cosine to analyze word frequency matrices and calculate distance matrices on Document Term Matrix formed by LDA (Latent Dirichlet Allocation) to determine similarities between texts based on probabilistic models. The paper also explores supervised learning algorithms including K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Random Forest which are trained on a labeled corpus to predict the origin of fragments of spiritual literature from a given chapter with accuracy measured for effectiveness in prediction. Overall this research aims to create a corpus where document is smallest unit of data after data cleaning followed by generating Bag of Words DTM for analysis through various similarity measures and predictive aspects so as to confirm or discover some closeness among sacred texts.

• - The paper explores similarities between various sacred texts using text mining techniques
• - Sacred texts can vary based on factors such as geographical location or the time of the birth of a particular religion
• - Despite differences, there may be similarities in the lessons taught by these texts
• - The study uses Asian texts (Tao Te Ching, Buddhism, Yogasutra, Upanishad) and non-Asian texts (four Bible texts) as the corpus
• - Similarity is measured using different measures like Euclidean, Manhattan, Jaccard, and Cosine applied to raw Document Term Frequency (DTM) and normalized DTM
• - Supervised learning algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Random Forest are used to predict the correct sacred text for any given chapter in the corpus
• - K-means clustering visualizations on Euclidean distances of raw DTM reveal patterns of similarity among the sacred texts
• - Upanishads and Tao Te Ching are found to be the most similar texts in the corpus
• - The research aims to find similarities between various sacred texts in terms of what they teach and how they teach religious lessons
• - Text mining using machine learning and feature extraction is employed to identify patterns in document collections
• - Similarity measures such as Euclidean, Manhattan, Jaccard, and Cosine are applied to analyze word frequency matrices and calculate distance matrices on Document Term Matrix formed by LDA (Latent Dirichlet Allocation)
• - Supervised learning algorithms including KNN, SVM, and Random Forest are trained on a labeled corpus to predict the origin of fragments of spiritual literature with accuracy measured for effectiveness in prediction

This paper is about finding similarities between different sacred texts using a computer program. Sacred texts are important religious books. The researchers used Asian and non-Asian texts to study. They measured the similarity using different methods and used computer programs to predict which text a chapter belongs to. They found that Upanishads and Tao Te Ching are very similar. The goal of the research is to understand what these texts teach and how they teach it. They used special techniques to analyze the words in the texts and trained computers to make predictions." Definitions- Sacred texts: Important religious books - Corpus: A collection of texts used for study - Similarity: How much two things are alike - Measures: Different ways of measuring something - Algorithms: Step-by-step instructions for a computer program - Clustering visualizations: Showing patterns in groups of things

Exploring Similarities Between Sacred Texts Using Text Mining Techniques

Religious teachings are often found in sacred texts, which can vary based on geographical location or the time of the birth of a particular religion. Despite these differences, there may be similarities in the lessons taught by these texts. This paper explores how text mining techniques can be used to identify patterns and measure similarity between various sacred texts.

Data Collection and Preprocessing

The corpus used for this study consists of Asian texts (Tao Te Ching, Buddhism, Yogasutra, Upanishad) and non-Asian texts (four Bible texts). The data was preprocessed to remove punctuation marks and stopwords before being converted into Document Term Frequency (DTM). Latent Dirichlet Allocation (LDA) was then applied to generate a Bag of Words DTM for analysis.

Similarity Measures

Different measures such as Euclidean, Manhattan, Jaccard, and Cosine were applied to raw DTM and normalized DTM in order to measure similarity between documents. These measures reveal similarity based on word usage. K-means clustering visualizations were also generated using Euclidean distances of raw DTM which revealed a pattern of similarity among the sacred texts with Upanishads and Tao Te Ching being the most similar text in the corpus.

Supervised Learning Algorithms

In addition to measuring similarity between documents using distance matrices from LDA models, supervised learning algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Random Forest were used to predict the correct origin for any given chapter in the corpus. Accuracy was measured for effectiveness in prediction.

Conclusion

This research aimed at creating a corpus where document is smallest unit of data after data cleaning followed by generating Bag of Words DTM for analysis through various similarity measures and predictive aspects so as to confirm or discover some closeness among sacred texts. The results showed that there are similarities between different religious teachings across cultures when analyzed using text mining techniques like feature extraction methods combined with machine learning algorithms like KNN SVM etc., along with traditional distance metrics like Euclidean Manhattan Jaccard Cosine etc..