bank indonesia

A Real-time view on citizen's sentiment

A centralized, query-able API service and data archive of public opinion towards monetary policies across 100+ social media channels.

Challenges

As the central bank of Indonesia, Bank Indonesia (BI) oversee and uphold the stability of the financial condition in the country. To accomplish that, BI evaluates its policy-making and regulatory decisions against the general public’s attitude towards it. One means of achieving that is through social media. However, gleaning insights from various social media sites are both difficult and time-consuming, particularly since there are 100+ pages, sites, and associated channels to pull these data from.

At a Glance

At a Glance

Volume, Velocity and Variety of Big Data

Challenges

Huge number of Bank Indonesia’s associated social media accounts (100+) with new data streaming in hourly (eg. facebook comments), and before the necessary ETL process these include duplicates, erroneous data that make it costly to store and efficient to process.

0 gb
Incoming data stream / hr
Silo-ed data making it difficult to correlate reactions from one social network to another, compounded by the fact that these channels — 121 in total — are widely varied in their data structures and schematic consistencies.
0
Active data streams

Centralized data warehousing as a service

Benefits​

0 %
Labour hours reduction

A centralized tracking infrastructure with a robust query-able API interface reducing the number of overhead work required for an analyst from 6 hours to 15 seconds (99.93% reduction)

Minimizing costly and time-consuming manual work in manually analyzing 100+ social media accounts and channels to ‘mine’ for insights on public sentiment, which represent a savings of $150,000 / annual
~$ 0
Annual Savings

Solutions

A proactive approach to social listening and sentiment monitoring from the Central Bank of Indonesia, powered by Supertype

Automated Pipeline & Data Warehousing Service

Supertype built an automated pipeline to scrape all social media data (Facebook, Instagram, Twitter, YouTube) that are on Bank Indonesia’s monitoring list, before cleaning, validating and storing the output on a private cloud storage (ETL)

Social Analytics Web Application

Supertype also developed a Next-based web application so any users can use a beautiful interface to filter and query for data:

  • Topic & category extraction of each post.
  • Regression analysis to identify key factors that affect public’s opinion, both in the positive and negative direction
  • Sentiment analysis of each comment to quantify a public reaction towards a certain regulatory detail or policy announcement

Bespoke API Service

In parallel, Supertype develop a token-based, authenticated API service to enable BI’s team to query, build upon, and analyze the data stored in the cloud storage for further analysis (e.g temporal, textual, engagement, sentiment analysis etc.) in a way that is secured, language-agnostic, and highly independent

English & Indonesian Sentiment Scoring Model

In addition to the API interface and web app interface, Supertype’s data scientist employ a sentiment scoring model that is robust to English and Indonesian (“Bahasa Indonesia”) language to account for the multi-lingual comments on social media sites

Enterprise-level access management & Admin

An admin portal was also developed so Bank Indonesia can easily moderate access to its data storage, provision new accounts for its web app, and implement administrative duties and permissions from a user-friendly interface

Interactive API documentation portal

An interactive API documentation portal that synchronizes with the data model was also created and presented to help foster a culture of ‘citizen data scientists’, where non-programmers can develop analysis without too much difficulty

Knowledge Transfer on social listening

Numerous consulting sessions and calls were carried out throughout the consulting period, culminating in a hands-on knowledge sharing session where Bank Indonesia gets to query the API following an in-person workshop

"Supertype has proven to be more than capable in translating our needs and expectations. Through their work we were able to visualize the social media analytics and citizen's sentiment. Additionally, Supertype ensures a smooth knowledge transfer process to our team (Bank Indonesia). "
wahyu_bi
Wahyu Indra Sukma
Digital Communication Team &
Performance Manager,

Bank Indonesia

Qualifying the value of Supertype’s data science & engineering work

Benefits & Key Impact

Minimize manual work

The ETL pipeline automatically retrieves internet comments from social media channels of BI's monitoring list and loads them into the centralized database periodically following a well-defined schematic process, saving an estimated 500 man-hours and $12,500 every month (~$150,000/year).

A uniform archive of public opinion data

Having a persistent single source of truth that combines data from more than 120 disparate sources periodically create a valuable archive of query-able data for BI's internal team of data scientists, social scientists and analysts

Comprehensive knowledge of public's reaction

The analysis of the social media data helps BI to better understand how public perceives monetary policies and relevant regulatory measures, and in turn, be more informed in its public communications effort and social campaigns

Contact Us

Want more case study?

This case study, along with others in our archive is also available as a PDF. To schedule a call with us and learn more about our past projects, leave your contact method below.

Your work email, so we can reach you. We respect your privacy and will not add you to any mailing list.

Inquiries

s@supertype.ai

More content from Supertype

Ă—

Table of Contents