Boxed In In Bangalore: Analyzing Sentiment On Indian Traffic Congestion

Click to enlarge. With a population of more than 1.2 billion, India is projected to be the world’s most populous country by 2025. By 2050, it is estimated that India’s urban population will constitute nearly half of that country’s total population, straining an already stressed infrastructure. The good news: Urbanization is an indicator of positive economic development. With improved urban planning, India can tackle urbanization challenges and increasing population to create a country that is poised for sustainable growth.

We heard a number of discussions about the potential for social listening intelligence last week at the Smarter Commerce Global Summit in Orlando.

This is an area I’ve been involved in within the IBM team for several years now, starting with some early explorations for how social data could be informative for our marketing efforts stretching all the way back to 2008.

It’s been exciting to watch this space evolve and mature, and with the advent of the IBM Social Sentiment index, we’re starting to see very practical uses of social data for better understanding if not the wisdom, then certainly the perspectives, of the crowd.

Yesterday, IBM held a Smarter Cities Forum in New Delhi, India, where we unveiled a new social sentiment capability to assist our customers in their Smarter Cities engagements.

We also unveiled findings from the latest IBM Social Sentiment Index on traffic, which looked at public sentiment across India’s largest cities — Bangalore, New Delhi and Mumbai.

Boxed In In Bangalore

If you’ve never experienced traffic in India, you can get a taste of the Sunday traffic in this video I shot during my first visit in June 2010.

But the recent analysis of publically available social media showed that the worst congestion in India is primarily caused by accidents and bad weather (three out of four times) when looking at the three cities together.

It also indicated some interesting variations between the three. For example, social conversation in Mumbai about stress around traffic is about half as high as Bangalore and New Delhi; references to the impact of rush hour on congestion in New Delhi are between five and seven times more negative than in Bangalore and Mumbai.

With a wealth of online content and public commentary on social channels such as Twitter and Facebook, city officials need new ways to measure positive, neutral and negative opinions shared by citizens regarding important city issues.

IBM’s advanced analytics and natural language processing technologies used to analyze large volumes of public social media data in order to assess and understand citizen opinions are now available to city governments around the world via new capabilities delivered with the IBM Intelligent Operations Center (IOC) for Smarter Cities.

Making Cities Smarter: The IBM Intelligent Operations Center

The IOC — which combines IBM software and services to integrate city operations through a single dashboard view to help cities improve efficiency — is now augmented with social media analytics capabilities that will help city officials make more informed decisions by looking at unfiltered citizen attitudes and actions, distinguishing between sincerity and sarcasm and even predicting trends as they surface online.

Combining the knowledge that population will rapidly increase in Bangalore, New Delhi and Mumbai in the coming years, with sentiment on commuters’ preferred mode of transportation, could help these cities more accurately plan for needed investments in transportation infrastructure and its potential impact.

City officials could also gauge where public awareness campaigns need to be administered to shift commuters to different modes of transport in order to alleviate growing traffic congestion.

The IBM Social Sentiment Index on transportation in India’s three largest cities surfaced several insights including:

• The top three factors impacting traffic congestion that citizens in each city talked about most online were diverse. Delhites chattered about public transportation, weather and the stress of commuting, while Bangaloreans show more concern for their overall driving experience, construction and parking issues, and Mumbaikars are talking about private transportation, accidents and pollution more often.
• Conversation in Bangalore around parking is viewed three times more negatively than in the other cities. Despite recent infrastructure improvements, less pollution and a solid public transit system, Delhites are experiencing a far higher amount of stress (50 percent) than those in Mumbai (29 percent) or Bangalore (34 percent). Most likely, this can be explained by an uptick in rallies and weather events this year, as well as the recent power outage.
• Surprisingly, sentiment on the topic of construction was relatively positive in Bangalore and New Delhi, and positive and negative sentiment on infrastructure in each was relatively even. Together, these may suggest that the transportation infrastructure improvements being made over the last two years in each city are beginning to positively impact citizens.
• Analysis shows that the relative negative sentiment for rush hour (35 percent) is one of the key drivers impacting traffic in New Delhi, which may explain why citizens talk about stress significantly more than commuters in Mumbai or Bangalore.

By applying analytics capabilities to the area of social media sentiment, organizations are able to better understand public opinions, and city officials can gain additional insights in order to draw logical conclusions about where they should focus their attentions and resources.

For example:

• Take Bangalore, the technology hub of India. Understanding that most commuters prefer private transportation despite negative sentiment around parking and construction may indicate that city officials should consider if it makes sense to advocate for more commuters to use mass transit and invest in infrastructure that will keep up with demand as more companies locate there.
• Since Dehlite’s indicate that public transportation is the preferred mode of transportation, city officials could use this insight to study which areas have high ridership and less road traffic and then implement similar actions in highly congested areas.
• In Mumbai, negative sentiment around traffic and weather at the peak of monsoon season (August) generated 5.5 times more chatter than in November. If the city could measure the fluctuation of public sentiment on these potential causes over time combined with specific weather data like rainfall or temperature, it might be able to better prepare to divert traffic during monsoon season or determine areas where a public safety campaign is needed.

“Like all rapidly growing cities across the world, there are infrastructure growing pains in many Indian cities,” said Guru Banavar, vice president and chief technology officer, Smarter Cities, IBM. “However, when city officials can factor public sentiment — positive, negative or otherwise — around city services like transportation, they can more quickly pinpoint and prioritize areas that are top of mind for their citizens. This could mean more targeted investment, improving a particular city service, more effective communication about a service that is offered, and even surfacing best practices and successful efforts that could be applied to other zones of a city.”

Methodology: IBM Cognos Consumer Insights And 168,000+ Discussions

Public social media content was analyzed by IBM Cognos Consumer Insight, which assessed 168,330 online discussions from September 2011 to September 2012 across social platforms including Twitter, Facebook, Blogs, Forums and News Sources and derived 54,234 High Value Snippets through a series of advanced filtration techniques for insight analysis.

The IBM Social Sentiment Index helps companies tap into consumer desires and make more informed decisions by looking at unfiltered consumer attitudes and actions, distinguishing between sincerity and sarcasm, and even predicting trends.

About the IBM Social Sentiment Index

The IBM Social Sentiment Index uses advanced analytics and natural language processing technologies to analyze large volumes of social media data in order to assess public opinions. The Index can identify and measure positive, negative and neutral sentiments shared in public forums such as Twitter, blogs, message boards and other social media, and provide quick insights into consumer conversations about issues, products and services.

Representing a new form of market research, social sentiment analyses offer organizations new insights that can help them better understand and respond to consumer trends. For more information about IBM Business Analytics go here.

You can also follow the conversation at #IBMIndex on Twitter.

For more information about IBM Smarter Cities go here, and follow the conversation at #smartercities on Twitter.