Question 1
Recently, deepfake technology has been making headlines. The latest iteration in computer imagery, deepfakes are created when artificial intelligence (AI) is programmed to replace one person’s likeness with another in recorded video. (Resource:https://www.businessinsider.com/guides/tech/what-is-deepfake).
Knowing the benefits brought by online social media data contents, your company, a startup AI company in Singapore decides to apply social media analytics to systematically analyse public opinions about Deepfake by using the relevant data from YouTube. You are part of the analytics team who is responsible for this analytical initiative including data collection, data processing, modelling and recommendations.
(a) You need to conduct research on YouTube. Discuss at least three (3) examples of social media data that could be collected from YouTube which are useful for your analyses. (Word cap: 100 words) (10 marks)
(b) Examine the social media collection methods and recommend the most suitable one to collect the data stated in Question 1(a). (25 marks)
(c) You are assigned the task of implementing application authentication (OAuth) in
Google Cloud for the project named MKT365. Please show your steps to get the Client
ID and Secrete for this project. (35 marks)
(d) Apply three (3) textual data pre-processing techniques that you consider appropriate to process the unstructured data from the YouTube Link
(https://www.youtube.com/watch?v=gLoI9hAX9dw&themeRefresh=1) and make
them analysable. You need to attach the example(s) to explain each of the data pre-
processing techniques. (15 marks)
(e) You are required to visualise the sentiment analysis trending results of public opinions on the Deepfake technology to the senior management in the past three years. Evaluate the different methods in this data visualisation task
