An ESRC internship led to the development of commercial software that can analyse social media comments about a company’s operations and products, revealing potential business risks.
- The internship developed methods and software tools to identify social-network comments that could threaten a company’s reputation and brands
- Mark McGlashan’s work has contributed to the extension of Cognizant’s operations into new software-product areas
- It has enabled Cognizant to roll out a key feature in its software platform, giving it a major competitive advantage over others in the market
- The research assisted the development of QuantEye, an intelligent-business app for mobile devices which will feature metrics and analytics for measuring consumer purchasing behaviours, trends and attitudes.
"The new system will enable the identification and tracking of risky users over time, and will assess and enable the analysis of online networks to which these users belong." (Dr Jai Ganesh, Director, SMAC Solutions, Cognizant)
About the research
Many companies have been quick to recognise the power of social media such as Facebook, LinkedIn and Twitter to promote their operations and services to potential customers. They often use social media to promote products, build customer relations and collect market intelligence.
But social networks also provide a platform for anyone to comment, and public dialogue may not always be positive for a company. Managing a company’s status or brand can be difficult; what people are saying and thinking is important, and can be potentially risky for the business.
To help address these risks, a three-month internship was carried out by Mark McGlashan, a senior research associate specialising in discourse analysis at the ESRC Centre for Corpus Approaches to Social Science (CASS). The internship took place at Cognizant Technology Solutions - an international provider of information technology, consulting and business-process outsourcing services based in Bangalore, India.
The project involved designing and building a system for the detection and tracking of users and online content that might prove ‘risky’ to a company’s reputation and brand, as well as monitor consumer interest to generate leads for sales. McGlashan devised algorithms and automated processes to identify and analyse online communications that are of interest to a company – especially the risks.
“I drew on my work at CASS about the way people express attitudes and make evaluations in the language they use - such as how people talk about desiring something, or making value judgements about what’s good or what’s bad,” explains Mark McGlashan.
The system identifies ‘risky’ content by identifying key verbs and adjectives in public postings. For example, if the word “despair” is used, the system assumes the person is not happy with the brand.
McGlashan’s research has contributed to the development of new software, potentially improving the way customers of Cognizant identify essential business decisions based on conversations on social networks.