What and where to read about algorithms for determining the interest (s) of users?

Question

Algorithms

Good day to all.
Maybe someone came across materials (books / articles / reviews) describing algorithms and / or approaches for determining user interests based on fuzzy data (electronic correspondence / phone calls (voice) / maybe, for example, based on the logs of some / circle of contacts / etc.)?
A lot of marketing information on this topic from the “Big Data is vital” series, it helped someone increase sales / convict employees of theft / improve logistics, etc.
But there is very little information on the algorithms used, methods, etc. Most of the mentions end with the technologies used (such as MapReduce, BigTable implementations, etc.). Regarding the algorithms themselves, I somehow did not find it. Can you tell me where to look?

Answer 1

I can recommend a book where similar questions are covered in examples in Python:
“We program the collective mind.” Toby Segaran - though I'm not quite
sure what you asked exactly ...

Answer 2

I'm doing a similar task, with the difference that preference analysis is for a dating service. The application side is to calculate the spammer, offers of various services ... i.e. whose purpose is NOT to get to know each other.
Each user has some indicators (we have more than 50 of them). Information is collected at each step of the user's stay on the site. Next, we scroll through the data through the statistics, and identify user clusters. Anomalous users are "neadexates", those whose distance in n-dimensional space is more than two medians of the cluster.
Clusters
_ D0%B0%D0%BD%D0%B0%D0%BB%D0%B8%D0%B7
www.statsoft.ru/home/textbook/modules/stcluan.html
www.nickart.spb.ru/analysis/cluster. php

Answer 3

I think the following book might also be of use to you: Algorithms for the Intelligent Internet. Advanced m...