Discover and read the best of Twitter Threads about #recsys
Most recents (5)
Very cool work @SpotifyResearch by @ZahraNazari1 et al. on cross-domain #recsys (using music tastes to recommend podcasts) #SIGIR2020. Audio, such a cool domain for #recsys and IR 😎 
These type of analyses are really great, @SpotifyResearch; studying model behavior between age/gender groups.
Would have been great to explore cross-domain #recsys in my previous gig at FDMG (recommending @BNR podcasts/radio for @FD_Nieuws readers, and the other way around? 😁). Too bad those audiences weren't linked (afaik)
#recsys dataset + paper: "MIND contains 1M users, 160k English news articles and 15.7M impression logs. Every news article contains rich textual content including title, abstract, body, category and entities. " msnews.github.io
"Each impression log contains the click events, non-clicked events and historical news click behaviors of this user before this impression."
Wow! So much data 😳. Awesome! Hope this sparks done interesting research in news recommendation.
Wow! So much data 😳. Awesome! Hope this sparks done interesting research in news recommendation.
With all that data we could've probably fully done our #UMAP2020 paper's analysis on a public dataset 😅 graus.nu/publications/b…
Stoked that our paper "Beyond Optimizing for Clicks: Incorporating Editorial Values in News Recommendation" with @dsflu and @anca_dmtrch is accepted as a FULL PAPER at #UMAP2020! I am particularly happy with this publication because... 👇 (1/4)
1️⃣ In our paper we show how you can align algorithm design across stakeholders (data scientists + journalists), by effectively modeling an editorial value (dynamicness) in a news recommender
2️⃣ we present (more) empirical proof that #recsys (can) offer(s) users more diverse, serendipitous, and dynamic articles compared to editorially curated lists, and hence (can) help in avoiding, not creating filter bubbles!
I just learned that @Randstad published their #AI principles last year: randstad.lu/en/artificial-…
Guidelines are helpful! At FD Mediagroep we spent some effort to figure out how and where (not) to apply AI in the context of #journalism (which tbh was a fun/important process).
Guidelines are helpful! At FD Mediagroep we spent some effort to figure out how and where (not) to apply AI in the context of #journalism (which tbh was a fun/important process).
Regarding our process of implementing AI at FD Mediagroep (in particular regarding algorithmic personalization/#recsys), we aim to be able to share some insights on that in the near future!
The work of @dietmarjannach has turned out to be an invaluable resource in our past few weeks of testing our @FD_Nieuws #recsys, often providing directly applicable insights and frameworks. e.g.: "Measuring the Business Value of Recommender Systems” arxiv.org/pdf/1908.08328…
@dietmarjannach @FD_Nieuws More examples? OK:
Measuring the impact of online personalisation: Past, present and future (2019): sciencedirect.com/science/articl…
Measuring the impact of online personalisation: Past, present and future (2019): sciencedirect.com/science/articl…
@dietmarjannach @FD_Nieuws Beyond Accuracy: Evaluating Recommender Systems by Coverage and Serendipity (2010): citeseerx.ist.psu.edu/viewdoc/downlo…
