Narrative Science Systems: A Review

TL;DR

This paper reviews the challenges and algorithms involved in automatic narration systems, highlighting the focus on statistical data and the limited role of subjective evaluation in current research.

Contribution

It provides a systematic review of existing narration systems, emphasizing the predominance of statistical data processing and the need for improved evaluation methods.

Findings

Most systems focus on statistical data narration.

Subjective evaluation by experts is limited.

Challenges include handling large volumes of live data.

Abstract

Automatic narration of events and entities is the need of the hour, especially when live reporting is critical and volume of information to be narrated is huge. This paper discusses the challenges in this context, along with the algorithms used to build such systems. From a systematic study, we can infer that most of the work done in this area is related to statistical data. It was also found that subjective evaluation or contribution of experts is also limited for narration context.