All Good Recommendations Come with an Explanation

On November 29, 2022, Hoang successfully defended his dissertation entitled “Mining Product Textual Data for Recommendation Explanations”.

While we were then coming out of post-Covid era, the defense was still conducted in a hybrid way, as one of the examiners was based in Australia.

Hoang explaining comparative recommendation explanation (pun intended)

Hoang’s thesis looked into various means for deriving explanations for recommendations from textual reviews. In particular, his first work on synthesizing an explanation from multiple reviews won a Distinguished Paper Award at IJCAI-20.

Subsequently, his thesis also discussed alternative types of explanations, such as comparative forms when a product’s explanation would be presented in the context of comparable product(s). Another piece of work incorporated product questions and answers, which tended to be objective, to enrich recommendations based on reviews, which tended to be opinionated and subjective.

First group dinner that Hoang joined at TungLok XiHe, Orchard Central

Hoang came from a big family. It is no wonder that ever since he joined the Preferred.AI family in January 2018, he also knew how to take care of others, including organising the running group.

Post-defense group celebration with dinner and drinks at the Naughty Chef, Shenton Way

And run indeed he did. He probably clocked the most kilometers of anyone during a PhD candidature. As Hady once remarked, “We all know what happens when Hoang sees a road. Not one for stopping, he just keeps going.”

Run Hoang, run!