Data and Marketing Insights
The Data and Marketing Insights research unit combines modern marketing techniques with novel methods from machine learning, data science, and big data analysis. From social media analysis, to a better understanding of target audience demographics, to machine-assisted decision making, associated researchers at Bocconi cover a wide range of topics.
Researchers use various techniques from natural language processing, computer vision, and data analytics to generate insights into common marketing questions: what do consumers talk about, where does the market go, how is my company positioned.
Sentiment analysis (or opinion mining) offers the chance to understand what aspects of a company are perceived positively or negatively by customers. These techniques can be applied to social media streams on a constant basis to help companies better manage their online presence.
Content analysis uses topic modeling to cut through the noise and the chatter in online communications, and to find the issues that drive engagement. Combined with other ways of text classification, they can help uncover unexplored market niches and chances to enter new markets.
Positioning maps can help pinpoint the public perception of a company among its followers (is it fun, eco-friendly, luxurious?), and help position it more clearly with respect to its competitors along a variety of axis.
Neural network-based techniques like word embeddings allow us to precisely capture the meaning of entire documents, and to visualize what is going on: are we speaking the language of our customers? Do we distinguish ourselves in the market?
Beyond this, DMI offers a variety of services to its members and supporters: seminars and workshops, joint projects, industry outreach, web-based analysis services, and much more.
Researchers from DMI are involved in developing ways to measure healthy conversations online, explore the public opinion towards public health policies, and help track the evolving landscape of political promises. Their research is published at prestigious venues in marketing, natural language processing, and social science.
They are applying, developing, and honing their skills in collaborations with international companies, such as Twitter or Swedish Riksbanken, and are in constant exchange with international experts in academia and industry.
- Aparna Garimella, Carmen Banea, Dirk Hovy, & Rada Mihalcea. 2019. Gender Bias in Part-of-Speech Tagging and Dependency Parsing Data. ACL.
- Twitter Health Metrics (2019–2021)
- Riksbankens Jubileumsfond (2020-2022) Mixed methods for analyzing political parties’ promises to voters during election campaigns
DMI provides an online tool that allows researchers to find the most descriptive terms for the object of their studies in text.