Dt:27 Feb 2018
Last Sunday when I was at a social gathering I met one of my classmates from my engineering college.He told me that he had read my blogs on about building Tableau vizzes & was particularly interested in my findings as he was working on similar reports with Microsoft's Power BI.While we continued talking on the subject , he mentioned that he has worked on building reports over data extracted from twitter feeds & asked me to check if its possible with Tableau.
With traditional BI tools , extracting data from the internet meant that you would have to spend a lot of time researching and exploring solutions to get the whole thing to work. But with some of the modern day visualization tools the opportunities are easier to explore & implement.I was reading up on the internet and I found out that QlikView & QlikSense can connect to twitter feeds .The parent company provides a connector and has made it available for free trial.I intend to try this out when I explore QlikSense in the months to come.
For Tableau I read up that Cdata provides some connectors(ODBC drivers) which can connect to twitter (Not sure if they work well with Tableau ).These connectors can also be used for traditional BI tools .However I could not find a place where I could get these connectors for free.Currently they go for at least 299$/year for desktop subscriptions.
The next best possible solution to get twitter feeds on Tableau was by using Tableau Junkie's web data connector (Hosted here http://tableaujunkie.com/post/119681578798/creating-a-twitter-web-data-connector ) .The guy also talks about building your own web data connector using Twitter API .I will try to build one for myself when I get some time.
For now I used Tableau Junkie's Webdata connector to pick up tweets with mentions for @narendramodi. Here is what I came up with at the end of this exploration.
Analyzing tweets with words like 'mission' addressed to @narendramodi.
Analyzing the frequency of tweets with mention of @narendramodi
This concludes Part- I of our discovery on twitter integration with Tableau.
It looks like we could conduct some sentiment analysis using this metadata. All you would require is some good experience with string SQL functions(& not R for now :p) .I will reserve that for advance analysis at a later stage.
Tomorrow I will publish Part-II where we discover twitter mentions of the Arsenal Football Club along with some geo tagging & will also explore Tableau Dashboards.
Till then good bye !!!
Note: This article is dedicated to Mr.M from MAK who inspired me to research this.Always remember MAK makes it possible
Credits : http://tableaujunkie.com/ The website has lot of content on BI which you may be interested to read.It also hosts quite a few web data connectors .
Last Sunday when I was at a social gathering I met one of my classmates from my engineering college.He told me that he had read my blogs on about building Tableau vizzes & was particularly interested in my findings as he was working on similar reports with Microsoft's Power BI.While we continued talking on the subject , he mentioned that he has worked on building reports over data extracted from twitter feeds & asked me to check if its possible with Tableau.
With traditional BI tools , extracting data from the internet meant that you would have to spend a lot of time researching and exploring solutions to get the whole thing to work. But with some of the modern day visualization tools the opportunities are easier to explore & implement.I was reading up on the internet and I found out that QlikView & QlikSense can connect to twitter feeds .The parent company provides a connector and has made it available for free trial.I intend to try this out when I explore QlikSense in the months to come.
For Tableau I read up that Cdata provides some connectors(ODBC drivers) which can connect to twitter (Not sure if they work well with Tableau ).These connectors can also be used for traditional BI tools .However I could not find a place where I could get these connectors for free.Currently they go for at least 299$/year for desktop subscriptions.
The next best possible solution to get twitter feeds on Tableau was by using Tableau Junkie's web data connector (Hosted here http://tableaujunkie.com/post/119681578798/creating-a-twitter-web-data-connector ) .The guy also talks about building your own web data connector using Twitter API .I will try to build one for myself when I get some time.
For now I used Tableau Junkie's Webdata connector to pick up tweets with mentions for @narendramodi. Here is what I came up with at the end of this exploration.
Analyzing tweets with words like 'mission' addressed to @narendramodi.
Analyzing the frequency of tweets with mention of @narendramodi
This concludes Part- I of our discovery on twitter integration with Tableau.
It looks like we could conduct some sentiment analysis using this metadata. All you would require is some good experience with string SQL functions(& not R for now :p) .I will reserve that for advance analysis at a later stage.
Tomorrow I will publish Part-II where we discover twitter mentions of the Arsenal Football Club along with some geo tagging & will also explore Tableau Dashboards.
Till then good bye !!!
Note: This article is dedicated to Mr.M from MAK who inspired me to research this.Always remember MAK makes it possible
Credits : http://tableaujunkie.com/ The website has lot of content on BI which you may be interested to read.It also hosts quite a few web data connectors .
Thanks for showing us that we can indeed connect to twitter feeds from tableau/qlik. And since you seem to be blogging on data visualization, will you be exploring R as well?
ReplyDeleteP.S. - Why Arsenal? :( I bet #Wengerout will have the most frequency among the tweets mentioning Arsenal
Thanks for the response .Would have appreciated if you hadn't been anonymous :-)
ReplyDeleteI will be exploring R at a later stage as right now I have very little knowledge about it.
As far as picking Arsenal goes.Its the team that I have followed a lot .And just for you I will try to perform an analysis on the hashtags :-)