Our first module....

They say programming is not a very big task. But we, as VTU students have not done much of application oriented programming and hence it is a big task!

We are all excited to begin the coding work from tomorrow. Our first task is defined below. We are going to write the modules separately, individually.... It may sound like redundant work but it is going to serve the purpose. We are all going to be acquainted with  Python programming skills and also help us to understand the problem statement better which in turn will help us come out with better ideas.

So here goes our task.....

"We need to consider an input text and scan it sentence-wise. Identify the complicated words and replace them with appropriate synonyms."

Sounds like child's play.. Doesn't it? But there are a lot of complications which needs to be simplified.

Points to ponder:

• After I retrieve each sentence, how do I recognize the complicated words? I can either consider the frequency count of each word and if the frequency count is below a threshold value then I mark it as a complicated word. Otherwise, I can eliminate the pronoun, preposition and adverbs in a sentence and consider only nouns, verbs and adjectives for replacement.

• How do I know the appropriate synonym for replacement? Before replacement of the synonym I check for the frequency of occurrence of that particular sentence using Google search API. The search results will give me several suggestions for the sentence. If any of the sentences in the suggestions have more frequency of occurrence than the former, I can replace it with that sentence. Or I can replace the words with the first 5 synonyms which WordNet offers and check for their validity using Google search API. 

• What happens when there are more than one complicated word in a sentence? If I follow the synonym replacement technique for each word, the complexity increases. Hence I can apply both the methods mentioned above. If I do not find any similar sentences which mean the same then I can go for synonym replacement of each word. 

• How do I use the Google search API for my application? Google search API usually returns all the results for the query, some of which may not hold relevance to us at all. Hence Google API must be tailored to return only those results which we are interested in. I can consider only the top 10 results returned by Google.

It is going to be a lot of fun implementing this idea!  The best part of being a part of a research project is that you don't know whether an idea can be implemented or not but you are completely free to think of new ones and nobody can stop you :-)