Gathering, Classifying and Visualizing Results from Social Surveys using OCR and Machine Learning Techniques

Abstract

Statistical representation of data as graphics is useful for making its interpretation easier, and for helping on decision making processes.When it comes to information that is obtained as result of social surveys, the problem is not only on presenting it, but on gathering the results by different means, and classifying them into categories, considering that for open questions, this is a process that is commonly done in a manual way.This paper is aimed on presenting a software tool for gathering data from social surveys relying on OCR technology or through a web application, classifying it using Machine Learning techniques, and presenting the resulting information in a mostly automatic way.For illustrative purposes, a case of study of a survey that was applied in Puebla, Mexico, in 2013 is considered.Quantitative results were obtained and reported for the classification process, and general user interfaces of the developed application are presented and described.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 1 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.