Representation Bias Identification

Identifying Insufficient Data Coverage for Ordinal Continuous-Valued Attributes

Abstract

Appropriate training data is a requirement for building good machine-learned models. In this project, we study the notion of coverage for ordinal and continuous-valued attributes, by formalizing the intuition that the learned model can accurately predict only at data points for which there are "enough" similar data points in the training data set. We develop an efficient algorithm to identify uncovered regions in low-dimensional attribute feature space, by making a connection to Voronoi diagrams. We also develop a randomized approximation algorithm for use in high-dimensional attribute space.

Publications to cite:

[1] Abolfazl Asudeh, Nima Shahbazi, Zhongjun Jin, H. V. Jagadish. Identifying Insufficient Data Coverage for Ordinal Continuous-Valued Attributes. SIGMOD, 2021, ACM.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

Prerequisites

What things you need to install the software and how to install them

• Maven 3.6
• Java 8 JDK

Installing (Console)

In console

mvn clean install

Installing (Eclipse)

In Eclipse or other IDE, all packages should be automatically installed once imported. ished

Running the tests

Explain how to run the automated tests for this system

Command line arguments

Command line arguments to use when running test scripts

Option	Descriptions	Has arguments	Allow multiple values
-a	selected attribute values	Yes	Yes
-e	epsilon values	Yes	Yes
-h	show help	No
-i	input dataset data file name	Yes	No
-k	k values	Yes	Yes
-n	number of query points	Yes	Yes
-o	if store test result in a file	No
-p	number of repeats	Yes	No
-phi	phi values	Yes	Yes
-r	rho values	Yes	Yes
-s	input dataset schema file name	Yes	No

Run tests from console

Accuracy Test

Format

mvn -e exec:java@accuracy -Dexec.args="{command-line-arguments}"

Example

mvn -e exec:java@accuracy -Dexec.args="-i data/iris.data -s data/iris.schema -a sepalLength sepalWidth petalLength -k 3 -r 0.05 0.1 0.15 -n 2000 -p 100 -e 0.1 0.2 -phi 0.1 0.2"

Efficiency Test

Format

mvn -e exec:java@accuracy -Dexec.args="{command-line-arguments}"

Example

mvn -e exec:java@efficiency -Dexec.args="-i data/iris.data -s data/iris.schema -a sepalLength sepalWidth -k 2 -r 0.05 0.1 0.15 -n 1000 2000 -p 100"

From Eclipse

In Eclipse or other IDE, run src/test/java/umichdb/coverage2/TestCoverageChecker.java

Built With

• Smile - Java-based Machine Learning Pacakge
• Maven - Dependency Management

Authors

• Zhongjun Jin
• Nima Shahbazi
• Abolfazl Asudeh

License

This project is licensed under the MIT License - see the LICENSE.md file for details