I apologize in advance for my mediocre English, but this will not affect the meaning of my comment (of course) ;)
Then:
very disappointed
the only thing worthy of note is the template models, that can be usable for an "ordered workflow" although still optimizable
1) no reference to statistical notions behind visual analisys of variable (for example if possibily interaction or relation between two or more variable is
statistically significant --> no parametrics or non parametrics tests are exposed..)
2) visualization techniques are very standard...(for example more rappresentative approaches like mosaic display, contingency tables, stratified
analysis , sieve diagrams, association plot, and many others, are NOT mentioned..that are basis for categorical data analysis)
3) then no reference to handle significant analysis of categorical variables.
4) no reference to correctly handle missing values (only hint to randomforest's "na.roughfix" method that is a very elementary approach if totally inadvisable)
5) no reference to regression, time series or others reported topics in "Aim and Scope of the book"..and their assumptions.
6) totally elementary presentation of classification strategies and modelling.
7) excessive exposition of performance measures (only for classification tasks) that in the most kaggle challengies are unused..(except ROC curve).
8) unnecessarily repeating code for train, validation and test dataset
9) no reference to serious tuning methods in dedicated libraries , no optimization strategies are mentioned..
10) no relevant resampling staregies which represent an important part of the Machine Learning are reported
11) classic classification algorithms are presentated, like 'rpart', 'randomforest' and 'xgboost' in a EXTREME elementary way..any tutorial on github, kaggle, or any online resource is much superior.
12) what makes me laugh is also a recommendation that author make in "Exercise section" of Ensemble chapter:.."try also deep Neural networks as algorithm..
good...Deep learning is a beast.. it can not be handled or mastered even in a dedicated books, and he says: try deep learning...without even mentioning the algorithm or extremely new relevant libraries like TensorFlow framework and Keras (its wrapper available for Python and R)
Loading your book clubs
There was a problem loading your book clubs. Please try again.
Not in a club? Learn more
Join or create book clubs
Choose books together
Track your books
Bring your club to Amazon Book Clubs, start a new book club and invite your friends to join, or find a club that’s right for you for free.
Download the free Kindle app and start reading Kindle books instantly on your smartphone, tablet, or computer - no Kindle device required. Learn more
Read instantly on your browser with Kindle for Web.
Using your mobile phone camera - scan the code below and download the Kindle app.
Flip to back Flip to front
 |
