(Log in to vote)
Feature Engineering For Machine Learning : Principles And Techniques For Data Scientists (2018)

Call Number 006.31/ZHENG,A

(0 holds on 1 copy)
LocationCall NumberItem Status
Adult Nonfiction006.31/ZHENG,AAvailable
Published: Beijing: O'Reilly, 2018
Edition:  First edition
Description:  xiii, 200 pages : illustrations ; 24 cm
ISBN/ISSN: 9781491953242, 1491953241, 9781491953242,
Language:  English

Feature engineering is a crucial step in the machine-learning pipeline, yet this topic is rarely examined on its own. With this practical book, youll learn techniques for extracting and transforming featuresthe numeric representations of raw datainto formats for machine-learning models. Each chapter guides you through a single data problem, such as how to represent text or image data. Together, these examples illustrate the main principles of feature engineering

The machine learning pipeline -- Fancy tricks with simple numbers -- Text data : flattening, filtering, and chunking -- The effects of feature scaling : from bag-of-words to Tf-Idf -- Categorical variables : counting eggs in the age of robotic chickens -- Dimensionality reduction : squashing the data pancake with PCA -- Nonlinear featurization via K-means model stacking -- Automating the featurizer : image feature extraction and deep learning -- Back to the feature : building an academic paper recommender -- Linear modeling and linear algebra basics Related Searches:
Machine learning
Data mining
Added--201806 anf

Additional Credits:
Casari, Amanda, author

Login to write a review of your own.

Login to add this item to your list.

Lists can be used to compile collections of items that you may be interested in checking out at a later date. You may also create public lists and share your favorites with other AHML customers.
No tags, currently.

Login to add tags.

To create a multiple word tag such as Science Fiction, enclose both words in quotes, like: "Science Fiction"

Critic Reviews

If your status is Confirmed Registration, your spot for the event is confirmed.

If registration for this event is full, you will be placed on a waiting list. Wait listed registrants are moved to the confirmed registration list (in the order of registration) when cancelations are received. You will receive an email notification if you are moved from the wait list to the confirmed registration list.

6.012 Patron-Generated Content

The Library offers various venues in which patrons can contribute content that is accessible to the public.  These include, but are not limited to, blogs, reviews, forums, and social tagging on the Library’s website and catalog.  Any instance in which a patron posts written or recorded content to any of the Library’s venues that are accessible to the public is considered “patron-generated content” and is subject to this policy.
By contributing patron-generated content, patrons grant the Library an irrevocable, royalty-free, worldwide, perpetual right and license to use, copy, modify, display, archive, distribute, reproduce and create derivative works based upon that content.
By submitting patron-generated content, patrons warrant they are the sole authors or that they have obtained all necessary permission associated with copyrights and trademarks to submit such content.
Patrons are liable for the opinions expressed and the accuracy of the information contained in the content they submit.  The Library assumes no responsibility for such content.
The Library reserves the right not to post submitted content or to remove patron-generated content for any reason, including but not limited to:
  • content that is profane, obscene, or pornographic;
  • content that is abusive, discriminatory or hateful on account of race, national origin, religion, age, gender, disability, or sexual orientation;
  • content that contains threats, personal attacks, or harassment;
  • content that contains solicitations or advertisements;
  • content that is invasive of another person’s privacy;
  • content that is unrelated to the discussion or venue in which it is posted;
  • content that is in violation of the Library’s Code of Conduct or any other Library policy