Social Analytics

Social data analysis, building a food and food influencer recommender, a must have for your food business!

Background

Singapore is an international food paradise with a billion-dollar food industry. The food industry in Singapore is worth S$8.3 billion in annual receipts and makes S$4.3 billion annual contribution to the country's GDP with 48,000 workers employed. According to the Ministry of Trade & Industry's (MTI) Food Industry Transformation Map, there is a 4.5% of productivity growth and 2,000 new jobs planned for 2020. This makes Singapore's food industry an attractive sector for budding entrepreneurs.

Problem

Due to the current COVID-19 pandemic, a lot of usual businesses eg. manufacturing, offices, are suffering. In order to help them, we want to identify a promising industry for them to start a new business.

Furthermore, as a result of the COVID-19 pandemic and in its aftermath, existing food businesses are hard pressed to reconsider their offerings and business strategy in order to amass enough sales to keep up or fall out of the race.

This can be done by:

1. Making personalised recommendations on trending food for businesses to sell
2. Making personalised recommendations on relevant influencers to engage in promoting their businesses and food

Brief Workflow

1. Instagram Scraping
2. Corpus Creation
3. Sentiment Analysis
4. Food/Cuisine and Influencers Analysis
5. Social Network Analysis

Outcomes

• Data visualisation dashboards for new and existing business owners who are interested in the food industry

• Social Network Graph

Dataset Used

• Instagram scrapes of posts from the top 12 food influencers in Singapore from 30th May to 30th June 2020

View project:

• Website
• Proposal Pitch
• Final Presentation
• Code
• Report
• Medium Article
• Poster

The Team

• Terence Neo
• Gracia Yuwono Kwantalalu
• Kok Jim Meng
• Christal Poon

Acknowledgements

We would like to thank the following parties without whom this project would not have been possible:

• Professor Kyong Jin Shim, for not sleeping, increasing the energy of the class always, and for being the best Prof in this module. We couldn't have asked for anyone better
• TA Qi Haodi and TA Ng Swee Hong, for their invaluable support and advice, and for being always avaliable to answer our questions
• Singapore Management University, for sponsoring AWS credits for us to run our code

References

• https://www.enterprisesg.gov.sg/industries/type/food-manufacturing/industry-profile
• https://en.wikipedia.org/wiki/Category:Asian_cuisine_by_country
• https://en.wikipedia.org/wiki/Category:Singaporean_cuisine
• https://en.wikipedia.org/wiki/Singaporean_cuisine
• https://monkeylearn.com/text-classification/
• https://www.analyticsvidhya.com/blog/2018/04/a-comprehensive-guide-to-understand-and-implement-text-classification-in-python/
• https://towardsdatascience.com/end-to-end-topic-modeling-in-python-latent-dirichlet-allocation-lda-35ce4ed6b3e0
• https://simpleanalytical.com/how-to-web-scrape-wikipedia-python-urllib-beautiful-soup-pandas
• https://towardsdatascience.com/named-entity-recognition-with-nltk-and-spacy-8c4a7d88e7da
• https://towardsdatascience.com/custom-named-entity-recognition-using-spacy-7140ebbb3718
• https://towardsdatascience.com/step-by-step-tutorial-web-scraping-wikipedia-with-beautifulsoup-48d7f2dfa52d