TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: See if a movie is happy or sad with sentiment analysis

58 pointsby eljsabout 8 years ago

13 comments

etjossemabout 8 years ago
One of the major problems with sentiment analysis is that it doesn&#x27;t pick up on irony or subtext very well. Relying on word choice alone, one such analysis pinned &quot;Fitter Happier&quot; as one of the happiest Radiohead songs.<p><a href="http:&#x2F;&#x2F;rcharlie.com&#x2F;2017-02-16-fitteR-happieR&#x2F;" rel="nofollow">http:&#x2F;&#x2F;rcharlie.com&#x2F;2017-02-16-fitteR-happieR&#x2F;</a><p>Edit: sorry, I misremembered - it actually handled &quot;True Love Waits&quot; pretty well. It did pick up &quot;Fitter Happier&quot; as one of the happiest, that&#x27;s what struck me as strange.
评论 #14275926 未加载
评论 #14275719 未加载
评论 #14275890 未加载
btbuildemabout 8 years ago
Interesting project, horrible presentation.<p>The colour gradient on the graph is confusing - does green mean happy, blue unhappy, and the movie changes tone over time? Axes of the graph are not labeled, what are we looking at? But, those concerns are secondary.<p>Bar graph seems a poor choice here, given the nature of the data. Given that there doesn&#x27;t seem to be any correlation with time, the order of phrases doesn&#x27;t seem important -- you could forgo the linear presentation, and display the distribution of the data instead.<p>For example, you could bucket the sentiment values (-3 to -2, -2 to -1, etc) and use a histogram to show the counts in each bucket. This would enable you to compare different movies (one histogram per movie).
评论 #14275438 未加载
aisoftengabout 8 years ago
I&#x27;m always a little wary of the robustness of sentiment analysis; in my experience, if you take the time to check sentiment analysis results sentence by sentence, you will find a high error rate.<p>I haven&#x27;t confirmed by looking at the source, but my suspicion is that either most sentiment analysis implementations are rule based or are not well tuned.<p>My go-to example is IBM Watson&#x27;s sentiment analysis service rating &quot;I hope you die&quot; as very positive because the sentence is categorized as &quot;hopeful&quot;. Which I suppose it is, technically, and perhaps this is a case where I&#x27;m expecting too much because recognizing this particular example as having a negative sentiment requires much human abstract reasoning and inference, but the example remains nonetheless because real-world language usage that isn&#x27;t dry and technical is rife with these sorts of linguistic usages.
eljsabout 8 years ago
OP here! I was watching the Stanford NLP classes a while back (<a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;playlist?list=PLiNErZ5Bus8qNxNsFZFkh-9_CzZRW9iH9" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;playlist?list=PLiNErZ5Bus8qNxNsFZFkh...</a>) and ended up trying the part about sentiment analysis on srt files. The way it works right is quite primitive but you can still see some &quot;trends&quot; on most movies. If there are any suggestions on how I could make this smarter I would love to hear them!
评论 #14275035 未加载
评论 #14273831 未加载
评论 #14274511 未加载
gh1about 8 years ago
I like it. What do you think of a recommendation engine which tries to recommend movies based on patterns of happiness and sadness?
snissnabout 8 years ago
Why does the graph go from green to blue?
评论 #14274516 未加载
zoidbabout 8 years ago
You should let people see what other people submitted although it&#x27;s pretty easy by fiddling with the URL. I like seeing the word patterns but I really wonder how well it can detect true sentiment, not an easy thing to do. <a href="https:&#x2F;&#x2F;www.crealdo.com&#x2F;story&#x2F;movie-sentiment&#x2F;movies&#x2F;13" rel="nofollow">https:&#x2F;&#x2F;www.crealdo.com&#x2F;story&#x2F;movie-sentiment&#x2F;movies&#x2F;13</a>
rodrigocoelhoabout 8 years ago
Also interesting: Character-to-Character Sentiment Analysis in Shakespeare’s Plays (PDF)<p><a href="http:&#x2F;&#x2F;www.aclweb.org&#x2F;anthology&#x2F;P13-2085" rel="nofollow">http:&#x2F;&#x2F;www.aclweb.org&#x2F;anthology&#x2F;P13-2085</a>
AznHisokaabout 8 years ago
What about a movie lile &quot;Brazil&quot;? would that be considered happy or sad?
评论 #14273201 未加载
评论 #14273885 未加载
blackflame7000about 8 years ago
It&#x27;s broken. My sentiment of La La Land was nowhere near that cheerful lol. In all serious though, its pretty cool idea for data analytics.
huangc10about 8 years ago
At the top, can you put a score or average of the analysis so we know if the overall movie is happy or sad? That number will be interesting :D
kingkawnabout 8 years ago
Horrifying.
opticalsabout 8 years ago
Make it available in French!