mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Owen <sro...@gmail.com>
Subject Re: Memory and Speed Questions for Item-Based-Recommender
Date Tue, 14 Jul 2009 10:03:23 GMT
On Tue, Jul 14, 2009 at 10:58 AM, Thomas Rewig<trewig@mufin.com> wrote:
> Because I need a UserSimilarity to precompute. Maybe I overlook a importent
> detail, but if I do it in that way to compute all on the fly in a item-based
> way:
>
>       *** "Pre"-Recommender ***
>       // set the model for the ItemItemMatrix
>      this.similarityModel = new MySQLJDBCDataModel(cPoolDS,
> preferenceTableSim, userIDColumnSim, itemIDColumnSim, preferenceColumnSim);
>       // set the "ItemSimilarity" for the ItemItemMatrix
>      this.similarityItemSimilarity = new
> EuclideanDistanceSimilarity(this.similarityModel);
>       // set CaschingSimilarity
>      this.cachingItemSimilarity = new
> CachingItemSimilarity(this.similarityItemSimilarity, this.similarityModel);
>       *** Recommender ***
>      // set the model for the Recommender
>      this.model = new MySQLJDBCDataModel(cPoolDS, preferenceTable,
> userIDColumn, itemIDColumn, preferenceColumn);
>      // set the Recommender with the *cachingItemSimilarity*
>      this.recommender = new GenericUserBasedRecommender(this.model,
> this.cachingItemSimilarity);

But why are you using a user-based recommender here? I thought you
were using an item-based recommender, in the end, to produce actual
recommendations. Yes, of course you do not plug in an item-similarity
metric into a user-based recommender.

User GenericItemBasedRecommender.

Mime
View raw message