#recommendationsystems — Public Fediverse posts

The EU Slapped Temu With a $232 Million Fine. And It Just Got Started

The EU Slapped Temu With a $232 Million Fine. And It Just Got Started – Moby THE GIST…
#Europe #EU #EuropeanCommission #DSA #EUCommission #illegalproducts #onlineretailer #phonechargers #recommendationsystems #Temu #theEU #Theinvestigation
https://www.europesays.com/europe/55058/

Optimizing Recommendation Systems with JDK's Vector API

https://netflixtechblog.com/optimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec

#HackerNews #OptimizingRecommendationSystems #VectorAPI #JDK #TechBlog #NetflixTech #RecommendationSystems

Optimizing Recommendation Systems with JDK's Vector API

https://netflixtechblog.com/optimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec

#HackerNews #OptimizingRecommendationSystems #VectorAPI #JDK #TechBlog #NetflixTech #RecommendationSystems

Optimizing Recommendation Systems with JDK's Vector API

https://netflixtechblog.com/optimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec

#HackerNews #OptimizingRecommendationSystems #VectorAPI #JDK #TechBlog #NetflixTech #RecommendationSystems

Optimizing Recommendation Systems with JDK's Vector API

https://netflixtechblog.com/optimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec

#HackerNews #OptimizingRecommendationSystems #VectorAPI #JDK #TechBlog #NetflixTech #RecommendationSystems

Optimizing Recommendation Systems with JDK's Vector API

https://netflixtechblog.com/optimizing-recommendation-systems-with-jdks-vector-api-30d2830401ec

#HackerNews #OptimizingRecommendationSystems #VectorAPI #JDK #TechBlog #NetflixTech #RecommendationSystems

Search Engine Land: AI recommendation lists repeat less than 1% of the time: Study. “When ChatGPT, Claude, or Google’s AI get asked for brand or product recommendations, they almost never return the same list twice — and almost never in the same order. That’s the big finding from a new study from Rand Fishkin, CEO and co-founder of SparkToro, and Patrick O’Donnell, CTO and co-founder of […]

Counterfactual evaluation for recommendation systems

https://eugeneyan.com/writing/counterfactual-evaluation/

#HackerNews #CounterfactualEvaluation #RecommendationSystems #MachineLearning #DataScience #ArtificialIntelligence

WordPress: Let’s Grow Together: Introducing Recommended Blogs . “When you find a blog you genuinely enjoy, you can add it to your personal recommendations list. Your subscribers and readers can then see these recommendations when they visit your profile in the Reader or hover over your gravatar anywhere in the Reader.”

https://rbfirehose.com/2025/11/09/lets-grow-together-introducing-recommended-blogs-wordpress/

So ever since my table saw works again thanks to friends I met on the #Fedi, I'm back looking at woodworking youtubers. I do enjoy getting the little tips and tricks people use for calibrating, etc.

Right this second I'm about to build a real long tapering jig in fact, so's I can fix our backdoor, which has been getting progressively more fuct by the year. I digress.

Which leads us to, I just saw a video with the title, "Why didn't I know this before? I couldn't believe it myself" or something like that - zero information about the topic, and the thumbnail was a few planks laid out in a corner that would not fit neatly together.

There is the barest suggestion that you might learn something about making them fit, somewhere in there, but the title, friends, I cannot forgive that title. And so, into the Don't Recommend Channel dungeon he goes, along with all the poverty tourism channels, Usian politics channels, celebrities, etc.

I admit, my youtube suggestion feed is pretty monotonous, since I don't accept the slop they obviously are getting paid to put in front of me, and they won't help out anyone - no matter how high quality - who doesn't pay them. I would like to fix that. I would like to have a suggestion feed absolutely full of videos that were rated highly by real human users, with zero techbro interference.

Is there a third-party video recommendation system yet that just catalogues YT urls and topics and ratings, and offers up that database without commercial constraints/quotas?

Shall we get started on one?

Would you submit videos you believe are worthwhile to a system with vetted human curators?

We need to reinvent the old Yahoo directory, basically, but curated for relevance, Slop content, etc. It should be centred on the Fedi, and specifically, the extant blocklists and relevant include lists. We do pretty good for avoiding the Slop assault.

From there, a recommendation system that could act as a (unauthorized if nec) front end, allowing avoidance entirely of YT's feed. This recommendation system might also, if this is all AP-based, be adaptable into a Toot Recommendation engine as well.

#Fediverse #Youtube #recommendationsystems

Meta Partners with Arm to Power AI Personalization Systems with Energy-Efficient Data Center Chips https://www.byteseu.com/1470511/ #AIStock #ArmHoldings #ArmBased #ChipTechnology #Energy #MetaPlatforms #RecommendationSystems #SemiconductorCompanies

This article details PCIC’s deployment, A/B test lifts, virtual aisles impact, and future directions for combining category and item insights. https://hackernoon.com/aisles-of-the-future-how-pcics-category-item-blend-transforms-online-grocery-shopping #recommendationsystems

Knowledge-based recommendation systems are a powerful alternative to traditional collaborative filtering, especially in scenarios where user data is scarce or decisions are high-stakes (think buying a house or selecting industrial equipment). 🏠⚙️ These systems use domain knowledge and logical reasoning to provide precise, explainable recommendations. They excel in cold-start situations and complex domains with clear constraints. Want to learn more about how they work and when to use them? Check out this comprehensive guide: https://fanyangmeng.blog/knowledge-based-recommendation-systems/ #AI #Tech #RecommendationSystems

🚀 Dive deep into the riveting saga of #LLMs improving recommendation systems—because clearly, what we need is our algorithms to be more verbose and indecisive. 🤖🔍 Enjoy 43 minutes of linguistic gymnastics that could have been summarized in one tweet: "LLMs make recommendations wordier, not better." 🙄
https://eugeneyan.com/writing/recsys-llm/ #RecommendationSystems #AI #Linguistics #OverlyVerbose #HackerNews #ngated

(Recommendation Systems and Search) × LLMs

https://eugeneyan.com/writing/recsys-llm/

#HackerNews #RecommendationSystems #LLMs #MachineLearning #AIResearch #TechTrends

🚀 Getting into recommendation systems? Content-based filtering is a powerful approach that suggests items based on user preferences and item attributes. Unlike collaborative filtering, it doesn’t need a huge user base to work effectively!

🔍 Google’s ML guide breaks it down with key concepts & best practices: developers.google.com/machine-learning/recommendation

#MachineLearning #RecommendationSystems #AI

I am browsing Harry & David for a “thinking of you gift”. I think I’m seeing a deliberate pattern to upsell — every item page offers a couple suggestions of similar items that are about 25 to 50% more expensive. If the suggestions were based solely on customer shopping patterns, I’d expect occasionally to see a same-priced or less-expensive item, which is the case with most other online retailers.

#ScottThoughts #HarryAndDavid #RecommendationSystems

I am browsing Harry & David for a “thinking of you gift”. I think I’m seeing a deliberate pattern to upsell — every item page offers a couple suggestions of similar items that are about 25 to 50% more expensive. If the suggestions were based solely on customer shopping patterns, I’d expect occasionally to see a same-priced or less-expensive item, which is the case with most other online retailers.

#ScottThoughts #HarryAndDavid #RecommendationSystems

I am browsing Harry & David for a “thinking of you gift”. I think I’m seeing a deliberate pattern to upsell — every item page offers a couple suggestions of similar items that are about 25 to 50% more expensive. If the suggestions were based solely on customer shopping patterns, I’d expect occasionally to see a same-priced or less-expensive item, which is the case with most other online retailers.

#ScottThoughts #HarryAndDavid #RecommendationSystems

I am browsing Harry & David for a “thinking of you gift”. I think I’m seeing a deliberate pattern to upsell — every item page offers a couple suggestions of similar items that are about 25 to 50% more expensive. If the suggestions were based solely on customer shopping patterns, I’d expect occasionally to see a same-priced or less-expensive item, which is the case with most other online retailers.

#ScottThoughts #HarryAndDavid #RecommendationSystems

I am browsing Harry & David for a “thinking of you gift”. I think I’m seeing a deliberate pattern to upsell — every item page offers a couple suggestions of similar items that are about 25 to 50% more expensive. If the suggestions were based solely on customer shopping patterns, I’d expect occasionally to see a same-priced or less-expensive item, which is the case with most other online retailers.

#ScottThoughts #HarryAndDavid #RecommendationSystems

@lenhoang just published a video about TikTok's recommendation algorithms and its dangers.

It's quite a long one, but it is informative and well sourced.

🔗 https://www.youtube.com/watch?v=oOzYpKi99z4

#AI #ML #MachineLearning #RecSys #RecommendationSystems #TikTok #media