Decoding Information from noisy, redundant, and intentionally-distorted sources

TL;DR

This paper introduces a framework for decoding reliable information from large, noisy, and biased data sources, demonstrated through a voting system that ranks raters and ratees effectively.

Contribution

It presents a novel method for extracting accurate information from redundant and distorted data, including robustness against noise and intentional manipulation.

Findings

Effective decoding of information in noisy environments

Robustness against various types of data distortion

Simultaneous ranking of raters and ratees

Abstract

Advances in information technology reduce barriers to information propagation, but at the same time they also induce the information overload problem. For the making of various decisions, mere digestion of the relevant information has become a daunting task due to the massive amount of information available. This information, such as that generated by evaluation systems developed by various web sites, is in general useful but may be noisy and may also contain biased entries. In this study, we establish a framework to systematically tackle the challenging problem of information decoding in the presence of massive and redundant data. When applied to a voting system, our method simultaneously ranks the raters and the ratees using only the evaluation data, consisting of an array of scores each of which represents the rating of a ratee by a rater. Not only is our appraoch effective in decoding information, it is also shown to be robust against various hypothetical types of noise as well as intentional abuses.