Maytal Saar-Tsechansky
University of Texas at Austin
69 Papers
395 Citations
Maytal Saar-Tsechansky is an academic researcher from University of Texas at Austin. The author has contributed to research in topics: Computer science & Active learning. The author has an hindex of 17, co-authored 54 publications. Previous affiliations of Maytal Saar-Tsechansky include New York University.
Chat about Author
Papers
More Than Words: Quantifying Language to Measure Firms' Fundamentals
TL;DR: The authors examined whether a simple quantitative measure of language can be used to predict individual firms' accounting earnings and stock returns and found that the fraction of negative words in firm-specific news stories predicts low firm earnings.
More than Words: Quantifying Language to Measure Firms' Fundamentals
TL;DR: This paper examined whether a simple quantitative measure of language can be used to predict individual firms' accounting earnings and stock returns and found that the fraction of negative words in firm-specific news stories predicts low firm earnings.
1.8K
•Journal Article
Handling Missing Values when Applying Classification Models
TL;DR: A method for analyzing various components in a natural gas pipeline with the aid of a computer controlled gas chromatograph and the amount of components in the natural gas stream.
Active Sampling for Class Probability Estimation and Ranking
Maytal Saar-Tsechansky,Foster Provost +1 more
- 01 Feb 2004
TL;DR: In this paper, the authors present a sampling-based active learning method for estimating class probabilities and class-based rankings, which uses weighted sampling to account for a potential example's informative value for the rest of the input space.
Active feature-value acquisition for classifier induction
Prem Melville,Maytal Saar-Tsechansky,Foster Provost,Raymond J. Mooney +3 more
- 01 Nov 2004
TL;DR: This work presents an approach in which instances are selected for acquisition based on the current model's accuracy and its confidence in the prediction, which can induce accurate models using substantially fewer feature-value acquisitions as compared to alternative policies.