Domino Data Lab

JUNE 15, 2021

Over the next decade, the companies that will beat competitors will be “model-driven” businesses. These companies often undertake large data science efforts in order to shift from “data-driven” to “model-driven” operations, and to provide model-underpinned insights to the business.

Modeling 69

Modeling Data Science Data-driven Data Warehouse 69

The Unofficial Google Data Science Blog

JANUARY 14, 2016

by AMIR NAJMI Running live experiments on large-scale online services (LSOS) is an important aspect of data science. In this post we explore how and why we can be “ data-rich but information-poor ”. There are many reasons for the recent explosion of data and the resulting rise of data science.

Experimentation 41

Experimentation Statistics Metrics Measurement 41

The Unofficial Google Data Science Blog

JULY 22, 2020

In practice, one may want to use more complex models to make these estimates. For example, one may want to use a model that can pool the epoch estimates with each other via hierarchical modeling (a.k.a. These MAB algorithms are great at maximizing reward when the models are perfectly specified and probabilities are accurate.

Experimentation 95

Experimentation Statistics Testing Strategy 95

Domino Data Lab

MAY 20, 2021

In this article we discuss why fitting models on imbalanced datasets is problematic, and how class imbalance is typically addressed. Data mining for direct marketing: Problems and solutions. Proceedings of the Fourth International Conference on Knowledge Discovery and Data Mining, 73–79. References. link] Ling, C.

Machine Learning 136

Machine Learning Metrics Data mining Knowledge Discovery 136

The Unofficial Google Data Science Blog

NOVEMBER 4, 2015

But most common machine learning methods don’t give posteriors, and many don’t have explicit probability models. More precisely, our model is that $theta$ is drawn from a prior that depends on $t$, then $y$ comes from some known parametric family $f_theta$. Here, our items are query-ad pairs. Calculate posterior quantities of interest.

KDD 40

KDD Testing Machine Learning Modeling 40

Domino Data Lab

AUGUST 1, 2021

In this article we cover explainability for black-box models and show how to use different methods from the Skater framework to provide insights into the inner workings of a simple credit scoring neural network model. The interest in interpretation of machine learning has been rapidly accelerating in the last decade. See Ribeiro et al.

Modeling 139

Modeling Deep Learning Machine Learning Knowledge Discovery 139

The Unofficial Google Data Science Blog

FEBRUARY 29, 2016

Rare binary event example In the previous post , we discussed how rare binary events can be fundamental to the LSOS business model. Say we build a classifier to classify user sessions into two groups which we will call “dead” and “undead” to emphasize the importance of the rare purchase event to our business model.

Experimentation 40

Experimentation Metrics Statistics Measurement 40