I have a project in machine learning in which I need to analyze a curriculum vitae. for that I have to write a python program. It uses basic techniques of Natural Language Processing like word parsing,chunking,reg ex parser. If you run the algorithm you can easily capture information like name,email id,address,educational qualification,experience in seconds from a large number of documents. but I have a confusion between two methods LSTM (RNN) and NER? What do you think is the best method?
LSTM is a neural network architecture used for sequence prediction, whereas NER is name of NLP task.
The two are incomparable, as one is a method and the other type of problem.
The relationship between them is that LSTMs can can be used for sequence tagging, which can be used for NER.
I had worked on the same project for months. I want to say they both work well if you have enough labeled data. For those entities (likes: name,email id,address,educational qualification), Regular Express is enough good. For variance experiences, you need NER or DNN. If the number of date is small, NER is best.
You know that resume is semi-structured. And you can think the resume is combined by variance entities(likes: name, title, company, description text, date and so on), so you can combine them to expand your train dataset.
I would suggest you to use spacy and add your own custom labels for NER.