Dataset contains the features such as description, goal , category etc to predict the probability in decimal such as 0.1,0.8 etc,

Now in the next step need to find out the text words associated with high probability using RNN +LTSM or any other neural network

Please guide with the key steps or the reference sample

Regards, Swati

  • $\begingroup$ so you want to assess the importance of features (words/text). I think in LSTM text analysis this is not so easy. Please provide a clear description of your data and model. Otherwise it is hard to say what a solution can be. See here for som background: datascience.stackexchange.com/q/44644/71442 $\endgroup$
    – Peter
    Commented May 31, 2019 at 11:26
  • $\begingroup$ Thank You Peter, for the quick response !! Need to extract features in the Description text using RNN/using unsupervised learning and to identify features correlated (may be with feature importance) to success/failure (That indicates the success and failure values in the target variable).. $\endgroup$
    – jaiswati_b
    Commented May 31, 2019 at 11:49


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.