EssayCare

An automated way to grade your essay

Prerequisite

• Python (>= 3.7)

Usage

To install EssayCare and its dependencies, please run following commands:

$ git clone https://github.com/2020-CS372/automated-essay-assignment-grading.git    # Clone Repository
$ cd automated-assignment-grading
$ pip3 install -r requirements.txt                                                  # Install Dependencies

To use EssayCare, you should need CoreNLP to be installed. You can install by running following commands:

$ make setup        # Setup Stanford CoreNLP
$ make run-server   # Start CoreNLP Server

Once the server is running, you can run the EssayCare in another terminal.

Copy and Paste your essay to data/input.txt and then run the command,

$ python main.py score --corenlp-url=http://localhost:9000

You can also test EssayCare with our pre-prepared dataset.

$ python main.py sample --corenlp-url=http://localhost:9000

The sample runs with at most 10 essays. If you want to run with all essays, you can use all instead of sample.

Features

Semantics

• ambiguous: Detects ambiguous sentences.
• duplicates: Detects too frequently used words.

Syntax

• agreement: Detects wrong tense, plural/singular or third person verb.
• capitalization: Detects misplaced upper case or lower case letters.
• preposition: Detects misplaced or missing prepositions.
• punctuation: Detects misplaced or missing punctuations.
• sentence_length: Detects too long sentences.
• structure: Detects wrong sentence structure.
• typo: Detects misspelled words.

Plagiarism

• plagiarism: Detects plagiarism against data/plagiarism_data/source