Geek Answers Handbook

String matching using repeating neural networks

Recently, I began to study repetitive neural networks. So far I have been training a character level language model on tensorFlow using the Andrej Karpathy blog . It works great.

However, I could not find research on using RNN to match strings or identify keywords. For one of my projects, I need OCR of scanned documents, and then the converted text is analyzed for key data points. Most string matching methods do not include OCR conversion errors, resulting in a significant error.

Is it possible to teach RNN variations of the converted text that I receive and use it to search for keywords.

+7
string-matching machine-learning tensorflow recurrent-neural-network
source share

No one has answered this question yet.

See related questions:

7432
How to check if a string contains a substring in JavaScript?
2664
How to check if a string contains a specific word?
613
The role of bias in neural networks
369
The era against iteration in the training of neural networks
349
What are the advantages of artificial neural networks over vector support machines?
one
Are repeating neural networks a chain of neural networks?
0
Binary classification of recurrent neural networks
0
Tensorflow - Repeating Neural Networks
0
Understanding why RNN tensor flow does not learn these toys
0
bidirectional language model of character level in tensor flow

More articles:


All Articles