From 25935e3b3fdf20ae3f644bd38051602c86b66cb5 Mon Sep 17 00:00:00 2001 From: "hannes.kuchelmeister" Date: Fri, 31 Jan 2020 16:45:53 +0100 Subject: [PATCH] fix research gap section to be about content-based filtering --- 25_Outline/outline.tex | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/25_Outline/outline.tex b/25_Outline/outline.tex index 07ee8cb..83f972c 100644 --- a/25_Outline/outline.tex +++ b/25_Outline/outline.tex @@ -19,11 +19,9 @@ \section{Research Gap} -There exists research on group recommenders and research on recommenders for configuration but there does not exists research on recommendation for group configuration. An approach for group recommenders is collaborative filtering. This approach is used also in recommenders for configuration. That is why adapting these approaches to suit the use case of group recommenders for configuration. +There exists research on group recommenders and research on recommenders for configuration but there does not exists research on recommendation for group configuration. An approach for group recommenders is content-based filtering. This approach is used also in recommenders for configuration. That is why adapting this approaches to suit the use case of group recommenders for configuration will be analysed in this thesis. -Commonly for collaborative filtering with group recommenders the preferences of the group members are combined with an aggregation function to generate a group profile. This profile is compared to historic group profiles from other groups and their choice of items (e.g. songs to listen to, things to buy) is used for recommendations. - -Collaborative filtering in configuration on the other hand usually uses the similarity of the current unfinished configuration to historic configurations to give recommendations. +Commonly for content-based recommenders categories based on content are created and a separate user or group profile is generated based on the preferences of whole items. For configuration recommenders however this would create additional modelling or content grouping workload, therefore in this thesis it is proposed to use attributes of a configuration as distinguishing categories. \section{Recommender Systems}