There is no basis for machine learning. I'm asking you this question because I got confused after getting to know it.
I don't know how to categorize and define the ensemble models, such as Adaboost and XGBoost, and the text extraction techniques (?) bag of words and word2vec.
Techniques such as words and word2vec are different, but they are methods of extracting text and predicting the next text, right?
Adboost and XGBoost are models for prediction, but I don't have a clue what it's like.
They're both different categories, but they're the same in terms of predicting, but how should we define 'Adabost, XGBoos / bag of words, word2vec'?
I mean, it's kind of gibberish. When it comes to programming, Should 'Adaboust, XGBoost' be considered as the overall flow chart and 'bag of words, word2vec' as a function in it?
Should I understand that we implement models such as 'Adboost, XGBoost' using 'bag of words, word2vec' technique?
ai python machine-learning nlp
Machine learning algorithms, such as adabost and xgboost, are very common algorithms that, given the table-type training data (and the correct answer label), learn from them and classify them. You're going to have to find that algorithm and read it, but when you start real machine learning, you're going to think of it as a black box, and it's going to be enough.
Bow, word2vec, because text data is so-called unstructured data, you can think of it in a way that turns text into a table form. That is,
This is how you want data in a table form where each data has a number for the columns. Bow is to organize how many times each text has a word, and word2vec is a way to reduce the dimensionality (in hundreds and thousands) because there are too many columns in the bow format.
© 2024 OneMinuteCode. All rights reserved.