Answer the question
In order to leave comments, you need to log in
Which neural network algorithm is suitable?
Input data: foreign language dictionary. Each word has a date when it was added to the system
What needs to be predicted: the date by which 8000 words will be added to the system
Answer the question
In order to leave comments, you need to log in
Apparently any extrapolation algorithm will suit you better than a neural network.
A linear approximation of the number of words in your system for a given date.
We take and build a table of 2 columns:
1. Number of days from the beginning of adding words
2. Total number of words for this day.
We use the approximation by a linear function using the least squares method.
the first link from Yandex: www.cleverstudents.ru/articles/mnk.html
As a result, you get the following linear function:
y=A*t+B
y - number of words
t - time, days
Respectively, we get the days when you will have 8000 words:
t = (8000-B ) / A
I think here you understand that you do not need NA AT ALL.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question