My project is about creating a program that works as a grammar checker for Arab students of age 15+.
I’m not asking you to create a grammar checker from the scratch, but you will be given bellow the mistakes that needs to be corrected.
The mistakes are as follows:
1. Since Arab students are usually writing from right to left in the Arabic language, and the ordering of the adj-noun combination is the opposite of what is in the English language, students are more mistake maker in such a thing.
(For example: saying the sentence “My state ID is expired” is usually written and spoken by the beginners of Arab English language students as “My ID state is expired”.
Note: This point needs your attention as a programmer to the related adj-noun in a sentence while designing your way to fix such a mistake.
2. Arabic has no verb to be in the present tense, and no auxiliary do. Furthermore, there is a single present tense in Arabic, as compared to English, which has the simple and continuous forms. These differences result in errors such as She good teacher, When you come to Germany?, I flying to Egypt tomorrow or Where he going?
Moreover, having no modal verbs in Arabic. This, for example, leads to: From the possible that I am late. (I may be late.) Another common mistake is to infer that an
auxiliary is needed and make mistakes such as: Do I must do that?
You need to fix such problems if appeared in a sentence. (Pay your attention that this point is about many points in one but they’re gathered all together because they are done by Arabs for the same reason)
3. Work on the punctuation mistakes, the upper/lower case letters, and the a/an appearance in the sentences since Arabic make no distinction between upper and lower case letters and the rules of punctuation are much looser than in English.
For example: Karolyn was able to visit california carrying luggage but not with Kim’s daughter. This sentence should be corrected to be: Karolyn was able to visit California carrying a luggage, but not with Kim’s daughter.
4. The indefinite article does not exist in Arabic, leading to its omission when English requires it. There is a definite article but its use is not identical with the use of the definite article in English. In particular, Arab learners have problems with genitive constructions such as the boy’s dog. In Arabic this would be expressed as Dog the boy, which is how such constructions may be conveyed into English.
This is the end of the mistakes that needs to be corrected.
The system should work on Windows 8 (64-bit) software. Note that you need to use UGI, which is friendly for students of age 15+. Moreover, you need to notify the mistaken studentts with those three points:
1. The location of the mistake (or mistakes).
2. The correction of the mistake (or mistakes) in an appropriate way.
3. The grammatical rule that is related to his/her mistake (or mistakes, each rule with its related mistake).
Note that I was unable to do that in my program since I used nltk corpus only.
The nature of the program you will create should use a kind of rule-based code since nltk (or similar corpus) will not be enough to solve those points because the parts of speech were tagged manually by students previously and that isn’t enough to create the grammar checker which capable to solve the mentioned mistakes.
If you’re done with this job, I need you to write a report of the design you used Including Requirements Specifications, Your theory of what a successful solution to the problem would be like.
– what inputs needed
– what function would be provided
– by what measures the work could be evaluated.
Please do NOT hesitate to ask me if you would liked more clarification.