Here You Will Get Knowledge Of AI Based Automatic Answer Checker

Abstract

An automatic answer checkers of AI Based Automatic Answer Checker system that checks and marks typed answers similar to a human being. This software application is built to check subjective an and allocate marks to the user alter verifying the answer. The system stores the original answer so that the comparison could be made between the given answer and the correct answer. These answers are stored in a database. When a user takes the test he is provided with questions and area to type his answers. Once the user enters his/her he the system then compares this answer to original answer written in database and allocates marks accordingly. Both the answers need not be exactly same answers and allocates marks accordingly as good as a human being. word to word. The system consists of built-in artificial intelligence algorithm that verifies AI Based Automatic Answer Checker

Introduction

Examinations have always been part of every educational, non-educational organization Examinations are of two types – descriptive or objective. Every examination needs to be evaluated. Competitive examinations are of objective types, they are conducted on machine and evaluated on machine also. These systems are very useful in terms of saving resource. It has been observed that there is no provision to extend these systems to subjective questions. Due to some problems, theses systems cannot be used for evaluating subjective answers. Al automatic answer checker system will be able to check subjective answer typed by student in an examination. It will allocate marks accordingly as good as human being. It will help to reduce the wastage of resources like papers and much more things. It will also overcome human limitations and speed up the overall education system. Our main aim of the system is that it will be helpful in various universities in academics for conducting class test, unit test and final exams. AI Based Automatic Answer Checker

Motivation

There are about 3 million students enrolled in grades 13 through 16 in Pakistan’s 1,086 degree colleges and 161 universities, according to Pakistan Higher Education Commission report for 2013-14. The 3 million enrollments are 15% of the 20 million Pakistanis in the eligible age group of 18-24 years. On an average every institute has 4 examinations per year so when calculated more than around 6 million answer sheets are generated. Keeping in mind of faculties evaluating papers and assigning marks is a tedious job. The processes of evaluating the descriptive answer will not only reduce the usage of resources bit also overcome the human limitations. It will also speed up the overall educational system wait for a long tie for the reason. stem as the students won’t have to wait for a long time for the reason AI Based Automatic Answer Checker

Problem Statement:

An automatic subjective answer sheet checker that checks and marks written answer using in build Artificial Intelligence technologies that verify answer and allocate marks accordingly as good as human being.

Proposed System

As observed that automated checker systems usually contain only multiple choice questions and there was no provision to extend such systems to subjective domain. Our proposed system consists of in built artificial intelligence sensors that verify answers and allocate marks accordingly as good as a human being. in earlier system both answers need to be same word to word but in proposed systems answers need not to be same. On contrary, the user given answer should contain most of the words that are already stored as standard answer. The standard answer is stored in the database with the description keywords. The system will evaluate each answer by matching the keywords with the standard answer. The entire process consists of three main steps, keywords extraction, matching of keywords with standard answer, weighting the keywords & generating score. This evaluation system will grade the answers depending upon the number of keywords matched. AI Based Automatic Answer Checker

Why did we choose online system?

The need for online examination aroused mainly to overcome the drawbacks of the existing system. The main aim of the project is to ensure user friendly and more interactive software to the user. The online examination brings an easy interesting working environment, more clarity in presenting appropriate information to the user and also it gives faster access and retrieval of information from the database. Through this new system, the drawbacks that have been seen in the existing system can be neglected. Under this system questions can be stored and retrieved so that preparing of questions papers each time can be avoided. AI Based Automatic Answer Checker

Approaches:

There are various different approaches that can be followed in NLP i.e. Natural Languages Processing, to make smart systems in order to maximize the performance and efficiency.

Few of which includes,

Jaccard Similarity Index:

It is used in understanding the similarities between sample sets and produces the percentage similarity between the sets.

Cosine Similarity:

It is a metric used to measure how similar the documents are irrespective of their size.

As we talk about our system, we are using the Jaccard Similarity to find the percentage accuracy of the user’s input and the stored data. it is used in understanding the similarities between sample sets and produces the percentage similarity between the sets. AI Based Automatic Answer Checker

Algorithm

National Language Processing:

NLP is a field in machine learning with the ability of a computer to understand, analyze, manipulate, and potentially generate human language. We have used NLP in our system for text extraction.

(National Language Toolkit)NLTK:

NLTK is a popular open-source package in Python. Rather than building all tools from scratch, NLTK provides all common NLP Tasks.

Word Segmentation:

It involves dividing a large piece of continuous text into distinct units.

Sentence Breaking:

It involves placing sentence boundaries on a large piece of text.

Lemmatization:

Lemmatization is the process of grouping together the different inflected forms of a word so they can be analyzed as a single item. It reduces word variations to simpler forms that may help increase the coverage of NLP utilities. We will first perform lemmatization to reduce words to the same root word. In our case, “globe” and “globally” will both become ” “has” and “have” will both become “has”.

Jaccard Similarity Index:

The Jaccard index is defined as the size of the intersection of the two documents divided by the size of the union of the two documents.

The Jaccard Similarity Index can be used to represent the similarity between two documents. A value “0” means the documents are completely dissimilar, “1” that they are identical, and values between 0 and 1 representing a degree of similarity.

Breaking down the formula, the Jaccard Index is essentially the number in both sets, divided by the number in either set, multiplied by 100. This will produce a percentage measurement of similarity between the two sample sets.

Use a Bag of Words, which is the list of unique words in the document, with no frequency count.
A Bag of Words with Stop Words removed (common language words like “the”, “at” for English).
Sets that take into account the frequency of words in the document.

To calculate the Jaccard Similarity, the number of shared words intersection between vector.’ and vector2 is calculated, and divided by total number of unique words union. AI Based Automatic Answer Checker

Flow Of Project:

Methodology:

Smart examiner/answer checker is a system that checks answers automatically provided by the user. It decides whether the answer is right or wrong by comparing the user answer to the answer stored in the database. The main objective of this paper is to implement a smart system that improves performance factor and minimizes the human error while maximizing the accuracy.

The following steps are the part of the process of achieving the accurate results,

1.Input Answer:

For a given question supplied by the system, the user provides an answer. The answer must follow the length provided and must contain the words which are significant and will give a higher possibility of an accurate answer or result.

2. Using SQLite:

We are using SQLite to store our questions and answers that will be compared to the one provided by the user.

3. Detecting Keywords:

The system will detect significant words by comparing it with the inputted words. Next these words will be used for summarization and allocating marks.

4. Summarizing/Comparing:

With the help of the database, the significant words can be selected and an appropriate marking can be done.

5. Checking the correctness of an answer:

After summarization, if it truly matches the stored answer, it will be marked correct and marks will be given, if partially correct, few marks will be given but if it will be completely wrong, no marks shall be provided. AI Based Automatic Answer Checker

Functions And Libraries Used:

We are using Django framework for Python. The libraries and functions used are as follow

Paginator(): Django provides high-level and low-level ways to help you • manage paginated data — that is, data that’s split across several pages, with “Previous/Next” links.
Split(): It is a Python string method, returns a list of all the words in the string. By default, works by whitespace
Strip(): It is a Python string method that returns a copy of the string in which all chars have been stripped from the beginning and the end of the string (default whitespace characters).
Len(): It is a built-in function in python. You can use the len() to get the length of the given string, array, list, tuple, dictionary, etc. len helps provide the number of elements.
Upper(): The string upper() method converts all lowercase characters in a string into uppercase characters and returns it.

Manual Systems Drawbacks:

Manual systems are way behind in speed, performance and results in contrary to automatic systems. Such as.

Inconsistency in data entry, room for errors, miskeying information.
Large ongoing staff training cost.
System is dependent on good individuals
Reduction in sharing information and customer services.
Time consuming and costly to produce reports.
Lack of security.
Duplication of data entry.
Takes up a lot of space. The biggest downfall to manual document filing is the amount of space it can take up.
Prone to damage and being misplaced.
Hard to make changes.

Advantages

Examiners get bored and exhausted by checking many answer sheets, hence the system reduces their workload by automating the manual checking process accurately.
The system calculates the score and provides results instantly.
It removes human errors that commonly occur during manual checking and mark students on merit basis, mark them fairly.
The system provides an unbiased result.
This the system excludes human efforts and saves time and resources.
The proposed system evaluates the answer based on the keywords. By comparing the standard answer and the student’s answer marks is obtained if the student utilizes all the keywords mentioned in the standard answer. Hence the said system could be of great utility to the educators whenever they need to take a quick test for revision purpose, as it saves them the trouble of evaluating the bundle of papers

Future Enhancements

The ultimate goal of artificial intelligence is to create computer programs that can solve problems and achieve goals like humans would. There is scope in developing machines in robotics, computer vision, language detection machine, game playing, expert systems. speech recognition machine and much more.
In this project we have trained only two questions for testing purpose but we can also add more questions in it
Examinations play a very important role in in colleges, universities and various other educational institutes. Many educational institutes have their examinations conducted online. But these exams only contains multiple choice questions which are providing to be very efficient in testing the student’s aptitude, on the other hand fail to measure the conceptual knowledge a student or learner must possess. Therefore subjective answer must be included in online examinations.
We can make a quiz and objective type paper for class test as well.
We can also add WH question which restricts the user to answer in one or two lines.
We can also add Yes/No question, or true/False questions.
We can make a test with limited time duration and allow user to attempt it.
Grammar checking for more accuracy
We can add MCQs questions in it.
The relaxation rules can added according to different question.
The database can be upgraded so that the design can support a wide variety of answers.
The code will be modified so that it can answer the Questions with high precision and accuracy.
Also this system totally evaluating any circumstances of business.
In future we are planning to evaluate subjective answer with diagrams and mathematical expressions. The current system only evaluates answer written in English. Further it can be extended to evaluate answer written in other languages also. AI Based Automatic Answer Checker

CONCLUSION

It is concluded that Artificial Intelligence (AI) is the simulation of human intelligence by machines. In other words, it is the method by which machines demonstrate certain aspects of human intelligence like learning, reasoning and self- correction. Since its inception, AI has demonstrated unprecedented growth. The future of Artificial intelligence is hazy. But going by the bounds of progress AI has been making, it is clear AI will pervade every sphere of our life. Thus, a smart, user friendly automatic answering system with the ability of detecting and answering questions is developed. There are many answering systems which use the concept of natural language processing to answer the questions but they are not so accurate in retrieving the appropriate answer. The answering system explained in this paper has multiple improvements as compared to the systems mentioned above. In the system, users can enter the answer in the way they want. Thus the system is more efficient and less time consuming. AI Based Automatic Answer Checker

Future Of Ai In Business

How to automatically answer form responses with OpenAI and Zapier

Top 5 Amazing AI Based Automatic Answer Checker