1. A method for computational analysis system for themes of displacement and identity 1 in partition english literature, wherein the method comprising: 2 a thematic analysis framework, a preprocessing engine, and a corpus creation 3 module set up to handle digital literary texts. 4

2. The method as claimed in Claim 1, wherein the corpus construction module is 5 configured to collect, digitize, and standardize a diverse set of literary works related 6 to the Partition, including novels, short stories, memoirs, and oral narratives. 7

3. The method as claimed in Claim 1, wherein the preprocessing engine performs 8 tokenization, lemmatization, stop word removal, and sentence segmentation, while 9 maintaining culturally relevant and multilingual words. 10

4. The method as claimed in Claim 1, wherein the thematic analysis framework includes 11 techniques such as topic modeling, including Latent Dirichlet Allocation (LDA) and 12 Neural Topic Modeling, to identify recurring themes related to displacement and 13 identity.

Description:COMPUTATIONAL ANALYSIS SYSTEM FOR THEMES OF DISPLACEMENT AND
IDENTITY IN PARTITION ENGLISH LITERATURE
Technical Field
5
10
15
20
[0001] The embodiments herein generally relate to a method for computational
analysis system for themes of displacement and identity in partition English literature.
Description of the Related Art
[0002] The academic interest in the English literature of the Partition of India has
traditionally emphasized qualitative methods of close reading, informed by the tools of post
colonial studies. Though such methods provide rich readings of the texts, they are limited in
scope to uncover the macro-level thematic trends of large datasets.
[0003] The Digital Humanities tools of "distant reading," such as Voyant Tools and
MALLET, have been successfully applied to the task of genre classification and textuality.
However, the existing literature on the application of generic NLP tools to the task of text
analysis has mainly focused on Western canonical texts or on large diachronic datasets. The
specialized domains of historical trauma, such as the 1947 Partition of India, have been
critically under-explored.
[0004] Furthermore, the generic sentiment analysis tools have been found to be
inadequate to analyze the complex and often contradictory sentiment of the texts of the
Partition, where the nostalgia and the trauma often blend. There exists no special framework
to analyze the semantic domains of identity fragmentation and forced migration.
2
SUMMARY
[0005] In view of the foregoing, an embodiment herein provides a method for
computational analysis system for themes of displacement and identity in partition English
literature. In some embodiments, wherein a sophisticated computational tool, specifically
5
10
15
20
designed to unlock the complex relationship between themes of migration and identity in the
context of Partition English literature. In this context, the tool overcomes the limitations of
generic text processing tools by using a carefully constructed domain-specific dataset,
analyzed via a hybrid approach to Natural Language Processing.
[0006] The tool’s unique strength is its custom-designed semantic lexicon, which
identifies culturally unique terms of migration and trauma, commonly misinterpreted by
generic NLP tools. The tool’s unsupervised clustering mechanism analyzes the “geographies
of belonging,” visually connecting spatial terms with identity markers. Moreover, the tool’s
sentiment tracking module analyzes the temporal progression of character sentiment during
migration events, relating spatial fragmentation with psychological fragmentation. The tool’s
automated extraction and visualization of thematic elements provide a powerful tool for
literary scholars, analyzing the psychological effects of historical boundaries, converting
qualitative interpretation into quantitative knowledge.
[0007] These and other aspects of the embodiments herein will be better appreciated
and understood when considered in conjunction with the following description and the
accompanying drawings. It should be understood, however, that the following descriptions,
while indicating preferred embodiments and numerous specific details thereof, are given by
way of illustration and not of limitation. Many changes and modifications may be made within
3
the scope of the embodiments herein without departing from the spirit thereof, and the
embodiments herein include all such modifications.
BRIEF DESCRIPTION OF THE DRAWINGS
5
10
15
20
[0008] The embodiments herein will be better understood from the following detailed
description with reference to the drawings, in which:
[0009] FIG. 1 illustrates a method for computational analysis system for themes of
displacement and identity in partition English literature according to an embodiment herein.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
[0010] The embodiments herein and the various features and advantageous details
thereof are explained more fully with reference to the non-limiting embodiments that are
illustrated in the accompanying drawings and detailed in the following description.
Descriptions of well-known components and processing techniques are omitted so as to not
unnecessarily obscure the embodiments herein. The examples used herein are intended merely
to facilitate an understanding of ways in which the embodiments herein may be practiced and
to further enable those of skill in the art to practice the embodiments herein. Accordingly, the
examples should not be construed as limiting the scope of the embodiments herein.
[0011] FIG. 1 illustrates a method for computational analysis system for themes of
displacement and identity in partition English literature according to an embodiment herein. In
some embodiments, a method is predicated upon a multi-layered architectural framework that
converts unstructured literary texts into structured and quantified formats for analysis, enabling
detailed analysis and comprehension of how themes of displacement and identity are created
and developed in texts such as novels, short stories, and autobiographies. The proposed system
4
is based upon a pipeline architecture comprising four distinct technical components: Data
Ingestion and Pre-processing, Semantic Ontology Building, Contextual Analysis using NLP,
and Visual Analytics. This methodology is intended to ensure that the results produced by this
proposed system are grounded in literary theory and take advantage of the processing
5
10
15
20
capabilities offered by contemporary computing technology to recognize patterns that would
be impossible for human scholars to recognize.
[0012] The first part of the working methodology is Data Ingestion and Pre-processing,
which is a significant component in view of the stylistic differences in Partition literature. The
system is built to process a pre-curated dataset of digital texts, which may include works of
renowned authors such as Khushwant Singh and Bhisham Sahni, as well as lesser-known
diasporic texts. In terms of technical implementation, a text cleaning process is employed using
Python libraries such as NLTK and spaCy. Considering that Partition literature may include
code-switching, untranslated words such as qaum, refugee, and kafila, as well as archaic
spellings, a custom-built dictionary is employed in the pre-processing module. In addition,
Optical Character Recognition Post-Correction algorithms are employed to clean noise in
manuscripts. The text is further tokenized, lemmatized, and part-of-speech-tagged. However,
in view of the fact that Partition literature includes many proper nouns such as names of
characters, locations, and organizations, which are not addressed in a standard lemmatizer, a
Named Entity Recognition Model is employed to fine-tune a dataset of colonial and
postcolonial texts to obtain accurate data on characters, locations, and organizations.
[0013] In some embodiment, after the pre-processing, the process then moves on to the
development of a specific Semantic Ontology, which acts as the knowledge base of the system.
It is the technical aspect of the system where the abstract concepts of "Displacement" and
5
"Identity" are translated into computer language. The system follows a hybrid approach of
dictionary-based sentiment analysis and ontology-based semantic expansion. To understand
the concept of "Displacement," the system develops a semantic ontology of words related to
movement, loss of home, migration, crossing the border, etc. These include words like "flee,"
5
10
15
20
"refugee," "border," "caravan," "abandoned." To understand the concept of "Identity," the
system develops a semantic ontology of words related to religion, community, or psychological
states of self. These include words like "Hindu," "Muslim," "Sikh," "belonging," "alien,"
"minority." The technical solution follows the Word2Vec algorithm and the BERT
(Bidirectional Encoder Representations from Transformers) model, specifically trained on the
Partition corpus. Unlike other models, the specific training of the model on the Partition corpus
allows the system to understand the word "home" used in the Partition texts has a specific
semantic proximity to the word "loss" or "memory" rather than the word "comfort."
[0014] The third module carries out Contextual Analysis through sophisticated NLP
tools to measure the "frequency, intensity, and correlation" of these themes. It does so through
a two-pronged approach of Sentiment Analysis and Geospatial Extraction. For sentiment
analysis, the system uses a "Long Short-Term Memory" network, a Recurrent Neural Network
that can "learn long-term dependencies." This means that the system can "trace the emotional
development" of a character or a narrative. For example, the system can "algorithmically track"
the "transition" of a protagonist from a "rooted citizen" to a "displaced refugee" by "measuring
sentiment scores" of adjectives that define a character over a narrative. At the same time, the
system uses Geospatial Named Entity Recognition to analyze the physical theme of
displacement. It "extracts location-based words" to "generate 'trajectory maps' of movement."
This technological approach can help researchers "analyze the geography of Partition
6
literature" to "identify common paths of migration" or "liminal spaces" where "identity crises
are most acute." The system "calculates a 'Displacement Index' for each text," a "quantitative
measure" based on "density of movement verbs" and "spatial shift indicators."
[0015] In some embodiment, the methodology proposes an advanced solution for the
5
10
15
20
analysis of "Identity" through Social Network Analysis (SNA). In this regard, through the
analysis of character dialogues and interactions, character relationship graphs are generated.
In these graphs, each node refers to characters, while each edge refers to character interactions.
The novelty in this solution is the assignment of "Identity Weights" to each edge in the graph.
The system analyzes character language to identify markers of "othering" or bonding with
members from different communities. For instance, through techniques like cosine similarity
in vector space models, character language can be analyzed to determine how character speech
patterns adapt to or diverge from new social groups after displacement. The technical solution
provides empirical support for character identity fluidity, thus complicating monolithic notions
of community identity in Partition narratives.
[0016] In some embodiment, the method culminates in the creation of a Visual
Analytics Interface that is capable of interpreting and presenting the results obtained through
the computation. The technology stack for this layer includes the creation of a dashboard that
utilizes frameworks such as D3.js (Data-Driven Documents) or Tableau and is connected to a
backend database (PostgreSQL) that stores the processed vectors. The interface is capable of
presenting different views to the users. The "Macro Views" allow users to see thematic
prevalence over the entire corpus of text, whereas "Micro Views" enable users to zoom into
specific paragraphs or sentences. The key components of this interface include "Sentiment
Heatmaps" that can be overlaid on narrative timelines to identify peaks of trauma and
7
displacement, and "Identity Clusters" that can be used to identify the proximity of characters
to communal or psychological identity markers. The system also includes a "Theme Tracker"
that is capable of dynamically graphing the correlation between the theme of displacement and
identity. For instance, it can identify how a peak in physical displacement narrative can trigger
5
10
15
a linguistic shift in identity vocabulary.
[0017] The draft computational analysis system can be understood as an analytical
engine that brings together the gap between qualitative literary criticism and quantitative data
science. In this sense, it proposes a solution to the problems posed by the reading of trauma
and migration through a methodology that advances from pre-processing to semantic modeling
and network analysis. It does not substitute the literary scholar’s perspective, as it provides a
bird’s-eye view of the movement and formation of identities in the English literary canon of
Partition, while enhancing the literary scholar’s perspective through the integration of NLP,
machine learning, and visualization techniques, which allow for the successful mapping of the
intersection of geography and psychology in the unstructured chaos of Partition history and
literature. , Claims:I/We Claim:
1. A method for computational analysis system for themes of displacement and identity 1
in partition english literature, wherein the method comprising: 2
a thematic analysis framework, a preprocessing engine, and a corpus creation 3
module set up to handle digital literary texts. 4
2. The method as claimed in Claim 1, wherein the corpus construction module is 5
configured to collect, digitize, and standardize a diverse set of literary works related 6
to the Partition, including novels, short stories, memoirs, and oral narratives. 7
3. The method as claimed in Claim 1, wherein the preprocessing engine performs 8
tokenization, lemmatization, stop word removal, and sentence segmentation, while 9
maintaining culturally relevant and multilingual words. 10
4. The method as claimed in Claim 1, wherein the thematic analysis framework includes 11
techniques such as topic modeling, including Latent Dirichlet Allocation (LDA) and 12
Neural Topic Modeling, to identify recurring themes related to displacement and 13
identity.