Skip to content
/ GUI-CLI Public

GUI and CLI applications in Java for text analysis and data analysis tasks. The GUI application allows users to analyze text input, while the CLI program calculates basic statistics of data from a CSV file.

Notifications You must be signed in to change notification settings

tejb96/GUI-CLI

Repository files navigation

a5-analysis

A graphical and a command line user interface combining most of what we have seen.

This assignment has two mandatory parts:

and no optional part.

1. Text Analysis GUI

Problem Statement

Write a graphical user interface (GUI) application that allows a user to type or paste text into a text area. A button triggers measurement of number of sentences and words and these results are displayed in a separate text area within the GUI. Sentences are delimited by either period ., exclamation mark !, or question mark ?. Words are delimited by any whitespace. A second button allows clearing the input and output text areas.

GUI Specifications

The GUI should prompt the user to enter text in a JTextArea, the input. A second JTextArea is used to display the results, the output. There are two JButton elements: analyze and clear. The former to trigger text analysis and display of the results, the latter to clear all text areas.

Many of the components can be found in StatsGUI used in the previous assignment. You are free to layout the components, keeping in mind that multiple lines of text are potentially entered - the size of the input text area should allow a good view of the text.

Design Specifications

You need two Java classes for this project. The design follows StatsGUI and Stats classes in the previous example.

TextAnalysisGUI provides the GUI elements and implements ActionListener. Its constructor sets up the elements. This class has a main() where a TextAnalysisGUI is instantiated. When the analyze button is clicked, a new TextAnayzer object is instantiated with the text entered in the input area. The results are displayed in the output area by calling the TextAnayzer object's toString() method.

TextAnalyzer provides a constructor that takes a String text as argument. In the constructor, the number of sentences and number of words are calculated and stored in two private instance variables numOfSentences, numOfWords. The class has two getter methods for these instance variables. The method toString() is overriden to return a string representation containing the two instance variables and descriptive text.

Two files src/TextAnalysisGUI.java and src/TextAnalyzer.java are created to get you started.

Reporting

In the markdown file TextAnalysis.md add the UML class diagram of TextAnalysisGUI and TextAnalyzer, their dependencies and their relationships. Add JavaDoc and comments to all your classes. Include a screenshot demonstrating successful execution and outputs of analyzing the text in test.txt and feynman.txt. For the latter, there are 1071 words and 53 sentences.

2. Data Analysis

Problem Statement

Write a program that can calculate basic statistics of a single column of data read from a comma-separated values (csv) file.

A command line interface (CLI) program prompts the user for the csv file name. The first row in this file is the header row, following rows contain decimal numbers that can be represented as double. There are a maximum of 100 data rows. The number of columns may vary from file to file. Once the data is loaded, the header and the data are displayed. The user selects a column to be analyzed by entering the column index. Subsequently, the selected column header and values are printed along with basic statistics: min, max, sum and mean.

CLI Specifications

The dialog is already implemented in DataAnalysisCLI's run method. Input/Output is performed with an InteractiveCLI.

The dialog implemented in src/DataAnalysisCLI.java uses src/InteractiveCLI.java, both provided. Studying code in src/DataAnalysisCLI.java will be helpful to understand other classes and how they interact.

Design Specifications

In addition to DataAnalysisCLI, an interface IBasicStats is available in src/IBasicStats.java. This interface defines the basic statistics methods min, max, sum and mean to be implemented.

You need two additional Java classes for this project. Both classes use the strategy for storing data up to a maximum of 100 entries that was presented in Stats in the previous assignment.

DataFrame represents a spreadsheet of maximum 100 data rows and a variable number of columns depending on the csv file. Its constructor takes a file name of a properly formated csv file. The file is parsed line-by-line using Scanner. The constructor throws IOException. It saves the header row in an instance variable headerNames of type String array. A getter method getHeaderNames() returns headerNames. All data rows are stored in a 2D double array. A numOfRows instance variable keeps track of the number of data rows, numOfCols contains the number of columns. There are getters for both these instance variables. The class overrides toString() which returns a String representation of the header names and at most 10 data rows, one row per line. The public method getColumnByIndex(int) takes a valid index of a column and returns a DataSeries object with label equal to column header name, and containing all data values.

DataSeries represents a 1D collection of values. The class has two instance variables label (a String) and data (an array of double). Its constructor takes a String as parameter which is assigned to label, and it initializes the data array to a size of 100. An add() method allows adding values and a count instance variable keeps track of how many elements have been added. There is a getter method size() that returns count. The class overrides toString() which returns a String representation of the label and at most 10 data values on a single line. This class implements the IBasicStats interface. DataSeries is used by DataFrame and DataAnalysisCLI.

Reporting

In the markdown file DataAnalysis.md add the UML class diagram of DataAnalysisCLI, IBasicStats, DataFrame and DataSeries, their dependencies and their relationships. Add JavaDoc and comments to DataFrame and DataSeries. Include a screenshot demonstrating successful execution and outputs of analyzing a column in text.csv and diabetes100.csv. Verify your outputs of these two datasets with alternate means, e.g. excel. Report on your verification.

Reflection - Include a sentence or two about:

  • what you liked or disliked,
  • found interesting, confusing, challenging, motivating

while working on this assignment.

About

GUI and CLI applications in Java for text analysis and data analysis tasks. The GUI application allows users to analyze text input, while the CLI program calculates basic statistics of data from a CSV file.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages