My dataset is structured like this:
Product1 - Sentence1
Product2 - Sentence2
Product3 - Sentence3
. .
etc
The sentences look like this:
Product1 - "We suggest that you wear this stylish piece with gold-toned drop earrings, churidar leggings and flats to complete an understated look."
A possible question to this can be - "Do we suggest that you wear this stylish piece with gold-toned drop earrings, churidar leggings and flats to complete an understated look?" - This is what i get using http://www.cs.cmu.edu/~ark/mheilman/questions/
Q: How can I get an understated look?
Q: What can I wear with gold-toned drop earrings?
Q: What will give me a complete look with churidar leggings?
So, what i feel I want to do is - to create sub-question on the nouns/proNouns and adjectives in an semantically correct putting product in context.
Sentence classification is one of the simplest NLP tasks that have a wide range of applications including document classification, spam filtering, and sentiment analysis. Specifically, we’re going to look at the sentiment classifier and discuss its components in detail.
The first step in sentence classification is to represent variable-length sentences using neural networks. In this section, I’m going to present the concept of recurrent neural networks (RNNs), one of the most important concepts in deep NLP. Many modern NLP models use RNNs in some way.
Whereas, if you give just one keyword, the possibility for the model to generate a factually correct sentence is pretty low because of the fact that the keyword would have occurred under various contexts in the training data, leading to more confusion. — Intuition
The maximum number of questions will depend on the length of the input text, or more specifically the number of sentences and named entities containined within text. Note that the quality of some of the outputs will decrease for larger numbers of questions, as the QA Evaluator ranks generated questions and returns the best ones.
Please have a look at Google Cloud Natural Language API
It has 'Entity Recognition' which might help your purpose. Here are the some sample implementations.
I suggest you to take a look for natural language services on the internet. Sometimes ago I had a similar problem and I solved it by using a service that you can finde here.
On the website this is defined as:
"API.AI is a natural language understanding platform that makes it easy for developers (and non-developers) to design and integrate intelligent and sophisticated conversational user interfaces into mobile apps, web applications, devices, and bots."
The really cool thing about this is that you can train it with some
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With