The structure of the training dataset is as follows: variable 1- containing textual data, variable 2…n- containing numerical data. The test data would provide variable 1 (textual data) to the model, and other numerical variables need to be predicted.
Dear Nandita Chaudhuri ,
I hope I understand your question correctly,
if You want to predict numerical data in your model, then your problem is regression and there are several methods for implementing it, which one of them is CNN. also if you have input data variables with different type you need to standardize them first.
Dear Nandita Chaudhuri
These articles might be helpful, have a look:
1. https://towardsdatascience.com/machine-learning-basics-multiple-linear-regression-9c70f796e5e3
2. https://towardsdatascience.com/bert-roberta-distilbert-xlnet-which-one-to-use-3d5ab82ba5f8
3. https://www.geeksforgeeks.org/regression-classification-supervised-machine-learning/
Kind Regards
Qamar Ul Islam
Yes, you can use mutlivariate ststistical models, e.g., Matrix variate von-Misses Fisher distribution and matrix-variate distribution. Useful book is "Directional Statistic: [Directional Statistics - Google Books] OR Analysis of Multivariate Statistics by Murehead.
- Badges
- Science topic