NGordnet - Exploring Word Usage History

A browser-based tool for analyzing the history of word usage in English texts.

Overview

The NGordnet project was part of a comprehensive assignment for CS 61B. The goal was to create a browser-based tool for exploring the history of word usage in English texts. The front-end code, written in JavaScript and HTML, was provided, while I developed the back-end in Java to handle input and generate output for display. This project involved implementing various data structures and algorithms to manage and analyze large datasets efficiently.

Project Requirements

NGrams Viewer (2A):
The first phase focused on building an Ngrams viewer, which displays the historical frequency of individual words (1grams) over time using the Google Ngram dataset. Key components included:

  • TimeSeries: A custom extension of the TreeMap class to handle year-wise data points.
  • NGramMap: A class to interact with the Google Ngrams dataset, providing methods to retrieve word frequencies and compute relative frequencies over time.

Introduction Video

WordNet Integration (2B):
The second phase integrated the WordNet dataset, adding semantic analysis capabilities. Key tasks included:

  • HyponymsHandler: Handling requests to find hyponyms (specific instances) of given words using WordNet data.
  • Graph Operations: Developing a graph representation to manage and traverse semantic relationships in the WordNet dataset.
  • Combining Datasets: Synthesizing results from the NGramMap and WordNet data to provide comprehensive word analysis.

Introduction Video

Implementation

TimeSeries Class:

  • Extended the TreeMap class with Integer keys (years) and Double values (data points).
  • Implemented utility methods for year-wise operations such as addition, division, and data retrieval.

NGramMap Class:

  • Parsed large datasets to create maps of word counts and total counts over time.
  • Provided methods for retrieving word frequencies and computing relative frequencies.

HyponymsHandler Class:

  • Developed a graph-based approach to find and return hyponyms for given words.
  • Implemented methods to handle complex queries involving multiple words and year ranges.

Graph Operations:

  • Built a custom graph class to represent WordNet’s synset relationships.
  • Implemented efficient algorithms for graph traversal and data retrieval.

Web Integration:

  • Modified provided front-end code to interact with the back-end Java handlers.
  • Registered new handlers for different functionalities such as historical text and graphical plots.

Features

NGrams Viewer:
A tool to visualize the historical frequency of individual words over time using a subset of the Google Ngram dataset.

Dynamic Graph Handling:
Implemented a graph structure to manage and traverse WordNet’s semantic relationships efficiently.

Interactive Web Interface:
Developed handlers to process user queries and return results through an interactive web interface.

Conclusion

The NGordnet project was a multifaceted assignment that provided hands-on experience in data analysis, algorithm development, and web integration. By working on this project, I gained valuable skills in handling large datasets, developing efficient algorithms, and creating interactive web applications.