MKT365: Assume you are a social media marketing analyst in STB Your role regarding this new tourism development is to advise STB on appropriate social media: Social Media Metrics and Analytics Assignment, SUSS, Singapore

During the COVID-19 pandemic, Singapore tourism suffered a lot. Visitor arrivals fell by 85.7% in 2020 to just 2.7 million visitors. International visitor spending totalled S$4.4 billion between January and September 2020, a 78.4% decline from the same period in 2019.

With the reopening of Singapore to world travellers, the tourism sector is experiencing one of the fastest recovery amongst all the economic sectors in Singapore, so much so that in 2022, Singapore expects the tourism sector to achieve approximately 70% of pre-Covid-19 spending according.

To drive this, the Singapore Tourism Board (STB) has launched a Request for Proposals (RFP) to develop and operate an integrated tourism development at Jurong Lake District (JLD).

Assume you are a social media marketing analyst in STB. Your role regarding this new tourism development is to advise STB on appropriate social media strategy to create opportunities for investments in new tourism offerings.

The objective is to explore the use of social media to increase the online footprint of new tourism offerings which would lead to revenue growth. Also you are to recommend social media analytics techniques to assist STB in making decisions on social media strategies.

(a) To help the new tourism development at JLD attract more, discuss three (3)
opportunities that are enabled by establishing a social media presence and leveraging social media analytics.

(b) Examine the data types in this YouTube link as well as four (4) challenges of leveraging these types of social data for such new tourism business.

(c) Apply and implement the process of emotion analysis based on the data for tourism in Question 1(b).

(d) After realising that the performance of their existing API for sentiment analysis is inadequate, STB decided to collect real-world sentiment data and use it to train their customized sentiment analysis. The collected dataset has 1:90:9 class distribution (positive:neutral:negative sentiments). Recommend social media metrics that are relevant and useful to assess the sentiment classifier in this case.