3
$\begingroup$

I'm an almost graduated applied math student. I do some sporadic work in marketing. I have done a few market segmentation projects. I am soon going to do one which is important to me. I usually segment customer types: in a questionnaire people are asked about how they "feel" or what they "like" or consider "important" about a certain type of product or situation, and using this data (usually scaled from 1 to 5, 5 beeing a strong agreement with... or considering ... to be very important) I group people that respondend in a similar fashion. I have only used K-means clustering to achieve this.

This new project is somewhat different. This questionnaire is aimed towards "day moments" (breakfast - supper - dinner, for example) and people are asked at what time they consumed X and WHY? [questions like the ones explained before (scaled from 1 to 5, 5 beeing a strong agreement with... or considering ... to be very important) but in this context], so the idea here is to segment moments of consumption not consumers. My question is, has anybody heard of these analysis? and if so, what are people using to do this? Im sure I can use k-means as usual but I wanted to know what's happening in the world and what could be the pros and cons of studying a bit more and trying something new or just sticking to what I know.

Greetings to all.

$\endgroup$
1
$\begingroup$

What you're doing sounds like Experience Sampling. If you can get your hands on the actual paper, this one might be of use to you: http://psycnet.apa.org/psycinfo/1988-03173-001.

$\endgroup$
  • $\begingroup$ Any other ideas? I have been reading a bit on Experience Sampling and it sounds very interesting, however the people interested in this project already have in mind what wants to be done, I'm thinking more along the lines of how to choose one unsupervised learning technique over another to find patterns in the consumption of things throughout the day. Any ideas? @AlexanderMeasure $\endgroup$ – JEquihua Jun 9 '12 at 17:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.