TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: Which directions would you recommend to research?

4 点作者 Liu将近 14 年前
Hello,<p>I have recently started doing a PhD in Information Systems. My general research subjects are User Modeling/Information Filtering/Recommendation Systems. I will also try to look at those subjects from a mobile point of view.<p>Which "burning" topics/problems do you know that exists in this field? Which issues/directions would you recommend focusing on?<p>Thanks,<p>Liu

1 comment

maxdemarzi将近 14 年前
Filter Bubbles? <a href="http://www.ted.com/talks/eli_pariser_beware_online_filter_bubbles.html" rel="nofollow">http://www.ted.com/talks/eli_pariser_beware_online_filter_bu...</a><p>Feels a bit like over-fitting, getting stuck in a local maxima type of problem.<p>Take music. Why is it that people get stuck? I'm stuck. I feel it. 90's metal is all I listen to willingly (current British pop hitting my ears is GF's fault). There are "generations" of music and people just stop liking new songs. They feel unfamiliar.<p>Apply the same to a world view(news, politics) or anything really. How do we make recommendation systems that allow users to grow their tastes?