Hello Everyone,
This blog is assigned as part of the Digital Humanities Lab Session by Dilip Barad Sir, taught me the practical skill of using digital tools to analyze literature in a scientific way. I learned to load major works like those of Shakespeare and Jane Austen into Voyant Tools to quickly see their unique styles: the Word Cloud showed me Shakespeare's focus on power ('king', 'lord'), while the Links Graph proved Austen's focus on social titles ('mr', 'mrs'). Furthermore, using the CLiC tool and a special quiz, I gained the crucial ability to distinguish between authentic human creativity and algorithmic nonsense by analyzing the thematic coherence of the text, successfully demonstrating my capacity to use computational methods for literary interpretation.
Quiz on - "Were these poems written by a human or a computer?"
In this quiz, I learned:
I learned to carefully look at a poem's words to see if they made emotional and thematic sense, which is a sign of a human writer. If the words were correct but nonsensical or the images kept shifting strangely (like "A visage bounds and tumbles"), I learned that was a clue it was written by a computer program. My success in choosing "Computer" showed I learned to spot the difference between true artistic intent and algorithmic word mixing.
Voyant Activity
Learning Experience: Generating the Word Cloud
Corpus Selection and Upload: The experience begins with selecting a corpus heavily dominated by historical and dramatic language, specifically the Shakespeare's Plays corpus (as suggested by the vocabulary). You would start on the Voyant homepage, select "Open," and choose or upload this specific collection of texts.
Tool Navigation: Upon loading the corpus, the user experiences navigating the Voyant interface and identifying the Word Cloud (often called Cirrus in the Voyant toolbar) as the appropriate visualization for viewing frequency.
Frequency Interpretation: The core experience is understanding that the size of each word is directly proportional to its frequency in the entire corpus. This allows for immediate, visual interpretation.
Result: You observe that words related to authority and royalty ('king', 'lord', 'sir') and dramatic, commanding language ('shall', 'enter', 'com') are the largest.
Stopword Filtering (Potential Adjustment): The user learns that common, low-meaning words (like "the," "a," "is") must be filtered out (using a Stopword List in the tool settings) to reveal the thematically significant words. Without this step, the largest words would simply be basic function words.
Inferring Genre and Theme: The final learning experience is the ability to rapidly infer the genre and central themes of an unknown text collection. The resulting Word Cloud confirms that the corpus is rooted in drama and high-stakes themes, characterized by frequent use of terms for command, power, morality, and action ('shall', 'good', 'man').
Learning Experience: Generating the Links/Cirrus Visualization
Corpus Selection and Upload: The first experience is selecting and uploading the entire corpus of Jane Austen's novels (as implied by the book titles like Pride and Prejudice, Emma, Sense and Sensibility, etc., in the center). This establishes the dataset for the analysis.
Tool Selection: The user then experiences selecting the Links tool (or a similar co-occurrence tool) from the Voyant interface. This tool is designed to visualize the relationships between words and documents.
Keyword Entry and Filtering: The key experience is identifying and inputting a set of relevant keywords (e.g., 'mr', 'mrs', 'miss', 'title', 'good', 'know') into the tool's settings. The tool then calculates how frequently these words co-occur or appear in relation to the center nodes (the individual novels).
Interpretation of Visualization: The ultimate learning experience is interpreting the resulting star-like graph. The thick, prominent lines connecting words like 'mr', 'mrs', and 'miss' to nearly every novel demonstrate that social titles and address forms are a consistently high-frequency and unifying thematic element across the entire Austen corpus. This confirms the dominance of social discourse in her work.
Learning Experience
Tracking Thematic Development: The primary experience is learning to track the quantitative progression of key themes across the structure of a work, specifically Frankenstein (as indicated by the x-axis label). By following the individual lines (e.g., the high peak of 'man' in blue around segment 5 or the rise of 'shall' in orange later in the text), the user learns to identify the specific narrative segments where certain concepts (humanity, destiny/command, life, or family) become most prominent.
Comparative Distribution: The user learns to compare the frequency distributions of multiple words simultaneously. For example, observing how 'life' (green) and 'father' (pink) may peak at different points teaches the user to identify where the narrative focuses more on the creation process versus familial responsibility.
Visualizing Narrative Structure: This activity provides a powerful visual experience of how an author organizes and shifts focus throughout a text. The user learns that a seemingly flat, continuous narrative can be structurally analyzed into distinct, thematically varied sections based on its vocabulary.
Learning Experience
The experience involves actively performing a corpus search using the CLiC Concordance tool. By defining the search to look for the specific, concrete noun "chin" within a focused literary corpus (Austen's novels, as per instructions), the user learns the foundational method of keyword-in-context (KWIC) analysis. The resulting view allows the user to experience contextualizing data, seeing exactly how the searched word is used in a passage (like the highlighted instance shown). Furthermore, the user experiences applying subsets/filters (like selecting "Quotes"), demonstrating how to refine a search to investigate specific rhetorical elements, such as distinguishing between the narrator's language and the dialogue of characters. This provides an immediate, evidence-based understanding of the author's stylistic choices and thematic patterns related to physical description or character expression.
No comments:
Post a Comment