Team 20 : Election Candidate Profiling

Election Candidate Profiling



Motivation

India is the largest Democracy in the planet earth. Since 1947 we have conducted 17 general elections and numerous state / local body elections. Indian citizen how is over 18 years of old can vote if they have registered them self as Voters. Indian elections are most expensive event on earth, Over Rs 50000 Crores spend on 2019 general elections[1].
Irrespective of all modernization efforts still we do not have mechanism which allows voters to review and assess the candidature of candidates who are contesting in the Election . According to law and rules of the land, all candidates who wants to submit their application for contesting should submit the affidavit to the election officer. These affidavits contains declaration and numerous details of the candidate.
Since 2008 it became mandatory for Election commission of India to make affidavits public and available through their archives portal. 
 We wanted to build a system that can help voters to know and understand the candidates better , compare their peers and past candidates.


 


Why Knowing Candidate is crucial ?

We have conducted a preliminary survey through twitter where 59.6% of the people preferred voting on the basis of the party rather than the candidate. When asked about the importance of knowing the candidate before the voting, 92.7% voted for YES.

We observe majority of voters usually votes for party irrespective of which candidate is running for the post in their constituency. Elected candidate is going to represent their constituency and going to play very important role is echoing constituencies needs and concerns. Elected candidates plays crucial roles in passing bills and other major government decisions for half a decade.

Even though candidates declare their criminal and financial details, often these details are shadowed by campaigns , media support and corruption etc.





Why this is a CSS Problem ?

  • Election is most crucial social event and it decides course of nation/state/regions progress.
  • Polarization of political views , mass manipulation, artificial hype of candidate's capability by election campaigns. media houses coverage and bias often shadows voter's ability to pick right candidate for their constituency
  • Building a system to aid voters do due diligence before voting, gives power back to voters.
  • Any form of Positive nudge to voters to perform their duty will have positive impact on our society's governance. 

Challenges 

  • Affidavit Data in NIC archives are scanned copy of actual handwritten forms
  • All PDF contains data in regional language, there is no standard format , everyone types out affidavit in their own way.
  • No way to cross check errors in affidavit data.



Ethics and Privacy

Data collected for this project is already available in public portals and our analysis is built based only on Affidavit data and data from official sources. 

Constituency's statistics and other Meta data such as education , crime are collected from government agencies

All data given in affidavits are already made available in https://affidavit.eci.gov.in/ and https://affidavitarchive.eci.nic.in/ sites.


Data Collection

The data is collected from various sources:

  • Affidavit data (from myneta and lokdhaba by ashoka university )

    1. Criminal, Financial and Educational information of candidates who have contested, as also winners of elections to State Assemblies, the Parliament and a few local bodies
    2. Political Party Watch (PPW), started in 2008, data includes: Income-Expenditure Statement of various political parties (national, regional and unrecognized parties) 
    3. Donations above Rs 20,000 that are received by Political Parties (National & Regional)
  • National Crime Records Bureau (NCRB)

    1. The details regarding crime rates in districts, constituencies and major cities are collected from NCRB website, which is an Indian government agency responsible for collecting and analyzing crime data as defined by the Indian Penal Code (IPC) and Special and Local Laws (SLL).

    2. The details regarding different crimes committed are collected state wise, further district wise are as following: STATE/UT, DISTRICT, YEAR, MURDER, ATTEMPT TO MURDER, TOTAL IPC CRIMES etc. Since there was ambiguity in calculation of rate and rank in crimes, the total number of IPC cases are considered in each crime.

  • Census India
    1. The details regarding literacy rate are collected state wise and district wise from the Census India website. It contains the following parameters
    2. Schools present in a district by category (Govt aided, private, Having sanitation facilities, Having mid-day meal facilities etc)
    3. Students enrolled in the schools by category (male, female)
    4. Teachers enrolled in the schools by category (primary, upper primary, secondary)


Our approach 

  • Instead of building OCRs to read PDF files, we focused on collecting data through myneta.info and other lokdhaba portal. We performed data cleaning and curation into single dataset.
  • From Primary data , we have created derived dimensions that can help to interpret specific qualities of candidates
    • Assets Index
      • Assets = (assets of a person – min asset value)
    • Crime Index
      • Cases reported = cases + serious cases
      • Crime = 1-(max cases reported – cases of state) / (max cases reported – min cases)
    • Education Index
      • Education = minimum mean years of schooling in either district or state
    • Multi Dimensional index  

Outcomes

  1. Regional/Constituency wise trend data 
    District wise trend data across the nation

State wise trend data across the nation


With our data and computed metrics we could build regional trend data. the geo maps created shows the quality of the contestants by district and state wise

2. Exposing Data as a service


    1. As part of this project work , we also tried to create a Election data as a service. 
    2. Though this Data service , we can Democratizing election data by providing easy interface
    3. We have built a easy rest api interface using AWS Lambda and serverless framework
    4. This allows future researcher to consume the normalized election related data. and build their applications/research on top of this.




Possible future works :

  • Localization 
  • Combining more growth and impact related metrics - Welfare, infrastructure and other growth indicators
  • Disaster and pandemic recovery statistics 
  • Pivot information for Party
  • Combining Social data of candidate - Flagged/Soft moderated posts, Extreme speech content ratio. mentions in viral posts/topics. 

  
    

Summary 

  • Combining Multiple data source gives simpler ways to interpret candidature of candidates contesting in election
  • Curating basic candidate and constituency data , gives ability to voters to choose better representative 
  • District and state wise trend shows quality of politicians in the region
  • e.g. Candidate profile from North east region in India shows better candidate profiles 







Team Members:

  • ​Issac Balaji ​​
  • Karthikeyan Arumugam ​​
  • Arun Kumar Padakanti ​​
  • Dustakar Prasanth Rao ​​
  • Haridasu Yaswanth
  • Siddhant Chandra Kulshrestha ​​
  • Sourav Kumar Singh ​​
  • Vambaravelli Tharun Sai

References : 

[1] - https://cmsindia.org/sites/default/files/2019-05/Poll-Expenditure-the-2019-elections-cms-report.pdf 

[2] - https://affidavit.eci.gov.in/ - Election Commission official Website for Affidavits for current elections

[3] - https://affidavitarchive.eci.nic.in/- Election Commission official Website for Affidavits for all elections

[4] - https://www.myneta.info/ - Portal built by https://adrindia.org/ which contains translated/parsed affidavit data

[5] - https://lokdhaba.ashoka.edu.in/  - Lok Dhaba is a repository of Indian election results built by Ashoka University

[6] https://ncrb.gov.in/en - National Crime records bureau of India

 

Comments

Post a Comment

Popular posts from this blog

23: Understanding the discourses around CAA NRC

Image
About CAA - NRC The Citizenship (Amendment) Act, 2019 aims to facilitate grant of citizenship to migrants belonging to Hindu, Sikh, Buddhist, Jain, Parsi or Christian communities from Pakistan, Bangladesh and Afghanistan who entered into India on or before 31st December, 2014 (1). It was enacted by the Government of India on 12 December, 2019. This particular move by the government of India sparked nationwide protests against CAA and the National Register of Citizens (NRC). The NRC is the official record of all the legal citizens of India. The main reason for the protests was the exclusion of Muslims and other communities who fled from the same or neighboring countries along with refugees from Sri Lankan Tamils in India, Rohingyas from Myanmar, and Tibetan refugees (4).  Due to the widespread national and overseas unrest from this act, it became a very important event in Indian history. Therefore, this led to an extensive media coverage in India. Furthermore, detailed accounts of t...
Read more

Archaeological Data Analysis on Harappan Civilization

Image
  Introduction Deriving conclusions from historical datasets is one of the hardest problems historians have to solve. The problems include verifying information, deriving correlations, and defending conclusions. The complexity of the problem results in increased time for a solution.  Through this project, we have proposed an accurate and faster way of concluding new inferences from historical data. This methodology primarily has 4 parts 1. Collecting and preprocessing the dataset 2. Analysing data using modern machine learning techniques. 3. Validate results using existing Information 4. Derive conclusions To showcase the working of the proposed methodology we have designed an experiment, in which we have derived cultural and economical relationships between ancient cities of the Indus valley civilization. Project Description Experimentation Data Collection (https://github.com/sidgupta2205/CSS_Project/tree/descriptors/Data) In the first phase, we worked on research and knowled...
Read more

14 : Misinformation Spread in Social Networks

Image
  Misinformation Spread in Social Networks Problem Statement In the Information Age, social networking sites have become a notable agent for the spread of misinformation, fake news, and propaganda. Misinformation on social media spreads quickly in comparison to traditional media because of the lack of regulation and examination required before posting. These sites provide users with the capability to spread information quickly to other users without requiring the permission of a gatekeeper such as an editor who might otherwise require confirmation of the truth before allowing publication. A 2018 study of Twitter determined that, compared to accurate information, false information spread significantly faster, deeper, and more broadly.  In this study, we aim to better understand misinformation flows in social networks. Related Work The spread of misinformation has been an ongoing research topic for some time now but before understanding the spread of misinformation, we need a cl...
Read more