Content Based Filtering
Content Based Filtering and Collaborative Filtering are the technical basis for almost every Recommender System (German). Content Based Filtering measures recognizable parameters of an element and compares them with parameters from other elements.
Thereby, the recommender system can offer users products with characteristics and attributes similar to products he has watched in the shop before. Measures of similarity can be determined by key "properties". By counting matches or the frequency of keywords between two elements, recommendations can be operationalized. Individual parameters can also be weighted according to their relevance.
The difference between Content-Based Filterung and Guided Selling is that CBF compares products to find matches, while Guided Selling systems match user profils with products:
- Content-Based Filtering proposes products as alternative to the starting product through the comparison of content (for e.g. description text)
- Guided Selling advises users refering to their specific requirements and needs. It customizes and forms a user profil, which will be compared with products, to calculate the best possible product recommendation
Disadvantages of Content Based Filterings:
- CBF recommmends products very similar to previously viewed products. Therefore, the user only receives a very limited list of alternative products. It should be noted that CBF systems have problems in recommending elements that are less thematically linked
- The system can not give any information about usefulnessand quality of individual elements
or if it meets the requirements of the user (in case of functionality and pleasure)
- A problem can also exist due to the difference between the fixed set of parameters by the website operator and their content meaning for users.