• Corpus ID: 39991010

From Predictions to Data-Driven Decisions Using Machine Learning

  title={From Predictions to Data-Driven Decisions Using Machine Learning},
  author={Nathan Kallus},
Predictive analyses taking advantage of the recent explosion in the availability and accessibility of data have been made possible through flexible machine learning methodologies that are often well-suited to the variety and velocity of today’s data collection. This can be witnessed in recent works studying the predictive power of social media data and in the transformation of business practices around data. It is not clear, however, how to go from expected-value predictions based on predictive… 

Figures from this paper

A data-driven method for stochastic shortest path problem

The experimental results show that the proposed approach can achieve a comparatively high accuracy when the sampling size of travel time is large enough, and under some reasonable assumptions, the accuracy could be 100%.

Finding the Shortest Path in Stochastic Vehicle Routing: A Cardinality Minimization Approach

This paper reformulates the original shortest path problem as a cardinality minimization problem directly based on samples of travel time on each road link, which can be obtained from the GPS trajectory of vehicles, and applies an ℓ1-norm minimization technique and its variants to solve the cardinality problem.



The Big Data Newsvendor: Practical Insights from Machine Learning Analysis

Finite-sample performance bounds on the out-of-sample costs of the feature-based decisions shows the "Big Data" regime, when over-fitting dominates finite-sample bias, is defined by p > O(n-1/(2 8/p)}\sqrt{\log{(n)}}), and both regularized ML and KO are effective methods to handle over- fitting.

Classification and regression trees

  • W. Loh
  • Computer Science
    WIREs Data Mining Knowl. Discov.
  • 2011
This article gives an introduction to the subject of classification and regression trees by reviewing some widely available algorithms and comparing their capabilities, strengths, and weakness in two examples.

A decision-theoretic generalization of on-line learning and an application to boosting

The model studied can be interpreted as a broad, abstract extension of the well-studied on-line prediction model to a general decision-theoretic setting, and it is shown that the multiplicative weight-update Littlestone?Warmuth rule can be adapted to this model, yielding bounds that are slightly weaker in some cases, but applicable to a considerably more general class of learning problems.

Predicting consumer behavior with Web search

This work uses search query volume to forecast the opening weekend box-office revenue for feature films, first-month sales of video games, and the rank of songs on the Billboard Hot 100 chart, finding in all cases that search counts are highly predictive of future outcomes.

Optimization of conditional value-at risk

In an intensifying international competition banks are forced to place increased emphasis on enter-prise wide risk-/return management. Financial risks have to be limited and managed from a bank wide

Big data: the management revolution.

Big data, the authors write, is far more powerful than the analytics of the past. Executives can measure and therefore manage more precisely than ever before. They can make better predictions and

On the Complexity of Linear Prediction: Risk Bounds, Margin Bounds, and Regularization

This work characterizes the generalization ability of algorithms whose predictions are linear in the input vector. To this end, we provide sharp bounds for Rademacher and Gaussian complexities of

In Search of Attention

Turnover, extreme returns, news and advertising expense are indirect proxies of investor attention. In contrast, we propose a direct measure of investor demand for attention -- active attention --

Predicting crowd behavior with big public data

The study validates and quantifies the common intuition that data on social media (beyond mainstream news sources) are able to predict major events, in particular the 2013 Egyptian coup d'état.

On a universal strong law of large numbers for conditional expectations

A number of generalizations of the Kolmogorov strong law of large numbers are known including convex combinations of random variables (rvs) with random coefficients. In the case of pairs of i.i.d.