Show HN: London Feels – Sentiment-analysis of Londoners Tweets on a Map

I once did a map of the UK using sentiment analysis of the text of geotagged Flickr photos, hoping to find the areas which were more happier than others. Turned out there was no geographical pattern from that data.<p>Geographical analysis tools should be used in these types of analyses, apart from just looking at blobs on a map. I used k-means based cluster analysis to find groups of happy and sad areas but again the groups turned out to be nothing conclusive.<p>The web GIS company I ended up working for used sentiment analysis of tweets by aggregated them into regions, so as to find positive and negative areas during a specific timeframe (for example, US elections). The regions had demographics which could be used statistically, and in general some interesting patterns were observed.

Hi HN.<p>I&#x27;ve built a map that takes a geofenced stream of tweets and runs AFINN-111 sentiment analysis on them, and then displays them in real time on a map of London.<p>Negative sentiments are displayed as Red tweets, happy tweets are Blue.<p>The whole thing is built on node.js using node-tweet-stream, node-sentiment and socket.io. The frontend map is leaflet with stamen design&#x27;s Toner tiles.<p>It&#x27;s quite fun to watch, especially when there&#x27;s a football match or a concert. If you click on the &quot;follow tweets&quot; checkbox, new tweets pop up as they arrive, although currently that makes the map pan north.<p>Thanks!<p>Dario.

Very cool but before clicking on some dots I was wondering why everyone feels the same. The colors are not ideal for red&#x2F;green colorblind people (is it blue and purple?)<p>Maybe include a feature to select the colors for happy&#x2F;sad&#x2F;average with a button to return to defaults?<p>Black for sad, light grey for neutral, something like a medium bright green for happy would be my picks.

Cool! Hey, that&#x27;s very funny....<p>We did the same with Tweets and Surfing. <a href="http://devwax.herokuapp.com/" rel="nofollow">http:&#x2F;&#x2F;devwax.herokuapp.com&#x2F;</a> from the meetup: <a href="http://www.meetup.com/DevWax/" rel="nofollow">http:&#x2F;&#x2F;www.meetup.com&#x2F;DevWax&#x2F;</a>. It was all done in a weekend with some drinking and surfing, so it&#x27;s a bit rough. The trouble with surfing was that the locations are very disparate and hard to guess. Fun to have a go at though...<p>Are you in London? (We are)

Could perhaps be more accurately retitled &#x27;London claims to feel on social media&#x27; map. There&#x27;s a lot of literature examining how people present themselves in such venues and how it&#x27;s often an intentional communication (even if subconscious) to create a certain impression.

Neat site. As others have pointed out, the sentiment analysis is off in many cases. I&#x27;d be quite impressed if you managed to correctly classify this one, though: <a href="http://i.imgur.com/wmWUitu.jpg" rel="nofollow">http:&#x2F;&#x2F;i.imgur.com&#x2F;wmWUitu.jpg</a>

Nice site, although it does have issues with the analysis as outlined by others here.<p>One suggestion I would have is some sort of filtering based on the content of the tweets. This tweet returned &quot;feeling good&quot;:<p>&quot;New post featuring: @NewLookPRTeam @nextofficial @Matalan @hmunitedkingdom @uoeurope @Accessorize @ASOS <a href="http://t.co/nGsFj306xT" rel="nofollow">http:&#x2F;&#x2F;t.co&#x2F;nGsFj306xT</a> #fbloggers&quot;<p>This one returned &quot;feeling average&quot;:<p>&quot;@dannykobe17 @DanielRacheter @Khuds_ @shangambling @Umar_Wilshere19 @_mikenewell_ @BlueKay10 is that ollie?&quot;<p>Whereas there&#x27;s absolutely no real sentiment to derive from this sort of thing.

Good job, expect I find colors pretty unintuitive - why red means negative? It&#x27;s color definitly connected with love, anger, war etc. So blue is for cold or maybe not showing emotions. I would definitely rethink that. And senitment analysis not always work - get a tweet rated as &quot;sweet&quot; ended with &quot;:(&quot;.

Some of these are unintentionally hilarious without context. Here&#x27;s a real gem: <a href="https://imgur.com/c7Ly6Qm" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;c7Ly6Qm</a><p>All in all, though, impressive. Sure some are misclassified but it seems like a significant majority are not, including a lot of the hard ones. Good work!

I always have a look at CityDashboard : <a href="http://citydashboard.org/london/" rel="nofollow">http:&#x2F;&#x2F;citydashboard.org&#x2F;london&#x2F;</a><p>It&#x27;s got a bit more than just twitter feelz. Mostly Boris Bike usage rate :D

Clicked through ~20 of them and the analysis was completely off in most cases.

Cool idea. Unfortunately I&#x27;ve yet to see sentiment analysis even really come close to providing any useful insights. It&#x27;s just not accurate enough on 140 character tweets.

Yeah, the the realisation of it is nice enough to look at, but the sentiment analysis (as always for these kinds of things) is pretty wonky.

The dots all look exactly the same to me. I&#x27;m colorblind, which I&#x27;m sure is the reason for this.

A legend for what the colors mean would be extremely helpful.

Please make one for Manchester!