how to get age and gender of tweets

saba76

Hi guys
please help me
how can I get age and gender of tweets in rapid miner?
Details by search twitter doesnt satisfy me.

Telcontar120

There isn't a native RapidMiner operator to retrieve that information. You would have to see whether it is available via the Twitter API and create your own script to get it (assuming Twitter even publishes that information via the API).

saba76

thanks a lot

Telcontar120

There isn't a native RapidMiner operator to retrieve that information. You would have to see whether it is available via the Twitter API and create your own script to get it (assuming Twitter even publishes that information via the API).

saba76

thanks a lot

sgenzer

another somewhat sneaky method, if you can get the name of the user, is to use an onomastics engine like NamSor. There's a NamSor extension in the RapidMiner Marketplace.

Scott

saba76

@sgenzer wow that's interesting.
It can't be with user Id or user name? I mean it must be just with first name and last name?

sgenzer

@saba76 yes that's how onomastics works. I don't think you need both FN and LN but I would look at the NamSor's site. Elian Carsenat is the owner and he's a guru. Been doing this for years.

Another idea if you're clever and have the time...you could probably create some kind of ML pre-processing model that trains + scores twitter handles that would parse into FN and/or first initial + last name etc... For example, a human will probably guess (even without knowing me), that my twitter handle "@sgenzer" is a FirstinitialLastname combo. Our good friend @Telcontar120 on this community is twitter handle "@BrianTvenstrup" which any human will see as FirstnameLastname. Same with RapidMiner co-founder "@RalfKlinkenberg". Other co-founder Ingo Mierswa's handle is "@ingomierswa" which, I would guess, would fail most ML models. Sorry @IngoRM But you get the idea.

Hmm interesting challenge I've created...

saba76

@sgenzer oh Is this ML model ready on rapidminer?
well I have just 4 or 5 days for this project...

saba76

@sgenzer I need some items and just can extract subjectivity objectivity and negative positives about twitter texts. I need a little more information and don't know what to do. I need it for text mining.eg to text mine McDonald's on Twitter.

saba76

@sgenzer and one thing else! I want to clustrate but most of them include ids! I think it's because of retweets that are existed in texts and don't know how to ignore ids to have a good clustrating.

sgenzer

hello @saba76 sorry I have been off the past couple of days. Where are you now on your project?