Hi !
I'm a french student, and I'm working on Twitter for my master's dissertation. It's a sociological approach, and it's mainly qualitative (the main part of my researches is about interviews with Twitter's users in order to analyse their motivations, their background etc.).
But, in a first part, I want to use a more "quantitative" approach, using NodeXl. My research is about one specific community, with only few people. So I want to start with one specific twitter's user and I want to study all his network. So, I think NodeXL is accurate to do the job.
The problem is that all the documentation is in english and that I'm not really familiar with this kind of tool. I'have already test NodeXL + Gephi with few people and it seems quiet easy to use because I was testing, like a "try and see" way. But now I want to treat the "real" datas, I have several interrogations :
I want to get the whole network of my "user 0" followed and following, with the three edges for each, at a 1.5 level, with a limit of 150 people. -> my goal is to be able to create 3/4 groups like the political/professional/personal/cultural, and to distinguished the 15 persons my user 0 is the most connected to.
Does my configuration is relevant ? How long it will take ? 3 or 20 hours ?
Of course, there is the most common problem : the API. It's very long, and I've already try to extract this kind of "big data", but the software crashed. Can I find a way to make researches separately and then try to fusion data ? Without doubloon of course. And for example make the software running on different computers from different twitter accounts ?
And I'm really taker of advice to start.
Thx a lot !
Quentin
I'm a french student, and I'm working on Twitter for my master's dissertation. It's a sociological approach, and it's mainly qualitative (the main part of my researches is about interviews with Twitter's users in order to analyse their motivations, their background etc.).
But, in a first part, I want to use a more "quantitative" approach, using NodeXl. My research is about one specific community, with only few people. So I want to start with one specific twitter's user and I want to study all his network. So, I think NodeXL is accurate to do the job.
The problem is that all the documentation is in english and that I'm not really familiar with this kind of tool. I'have already test NodeXL + Gephi with few people and it seems quiet easy to use because I was testing, like a "try and see" way. But now I want to treat the "real" datas, I have several interrogations :
I want to get the whole network of my "user 0" followed and following, with the three edges for each, at a 1.5 level, with a limit of 150 people. -> my goal is to be able to create 3/4 groups like the political/professional/personal/cultural, and to distinguished the 15 persons my user 0 is the most connected to.
Does my configuration is relevant ? How long it will take ? 3 or 20 hours ?
Of course, there is the most common problem : the API. It's very long, and I've already try to extract this kind of "big data", but the software crashed. Can I find a way to make researches separately and then try to fusion data ? Without doubloon of course. And for example make the software running on different computers from different twitter accounts ?
And I'm really taker of advice to start.
Thx a lot !
Quentin