2014.ISDS.Abstracts.Final.pdf


ISDS Annual Conference Proceedings 2014. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, 
and reproduction in any medium, provided the original work is properly cited.

ISDS 2014 Conference Abstracts

Analytics, Machine Learning & NLP – use in 
BioSurveillance and Public Health practice
Mujitha B. K B*1, Ajil Jalal2, Vishnuprasad V1 and Nishad K A1

1Informatics, LongRiver Infotech, Bangalore, India; 2IIT Madras, Chennai, India

Objective
To summarize ways in which Analytics, Machine Learning (ML) 

and Natural Language Processing (NLP) can improve accuracy and 
efficiency in bio surveillance and public health practices. We also 
discuss the use of this framework in typical surveillance applications 
(Integration with Devices/Sensors, Web/Mobile, Clinical Records, 
Internet queries, Social/News media).

Introduction
Currently, there is an abundance of data coming from most of 

the surveillance environments and applications. Identification and 
filtering of responsive messages from this big data ocean and then 
processing these informative datasets to gain knowledge are the two 
real challenges in today’s applications.

Use of Analytics has revolutionized many areas. At LongRiver 
Infotech, we have used various Machine Learning techniques 
(Regression, Classification, Text Analytics, Decision Trees, 
Clustering etc.) in different types of applications. These methodologies 
are abstracted in a generic platform, which can be put to use in many 
public health and surveillance applications, which are enumerated 
here.

Methods
In this generic ML platform, we brought together modules covering 

each of the ML and NLP areas. This platform was then evaluated in a 
simulated environment – interfacing with a Web/mobile surveillance 
data capture application, medical devices/sensors over RFID, and 
social feeds. Surveillance data in both streaming and batch modes 
have been used for this test environment. ‘R’ was used for ML 
algorithms and Infrastructure tools like NoSQL database (Apache 
Spark), Map Reduce (Spark/Hadoop) and Visual tools (R/Tableau) 
were integrated in this pilot study.

Results
Each of the independent modules included in this environment had 

been evaluated in separate projects (Precision, Recall rates, R-squared 
values, AUC etc. for respective algorithms). Scaling capabilities 
(input data, ML processing) of the platform was evaluated in an 
Apache spark cluster.

Conclusions
This framework can be plugged into any surveillance application, 

which has the required IT infrastructure in place – for efficient and 
scalable distributed processing and big data handling.

From our evaluation so far, there is an increased interest from 
various stakeholders in using these Machine Learning algorithms and 
NLP technology on Surveillance data.

Further enhancements in NLP include:
1) Speech recognition, which enables users to tell their problems

(which can first be converted to text and then NLP can act upon it)
2) Support for multiple languages (which enables public to tell in

their own local language)
3) Question-Answering (which enables machine processing of user

stories and responding with the findings/solutions)

A primary motivation for presentation at this conference is to 
solicit feedback from public health practitioners on this idea and its 
potential / challenges for use in existing surveillance systems.

Analytics use in Bio Surveillance

Analytics use in Public Health practice

Keywords
responsive messages; clustering; classification; machine learning 
algorithm; text analytics

*Mujitha B. K B
E-mail: mujitha@longriverinfotech.com    

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * (1):e194, 201