What you'll find below it's a targeting system based on managed metadata content tagging and dinamically generated queries using XRANK directives.
Let's think about a enterprise model with a hierarchy like this
- Channel 1 (00000000-0000-0000-0000-000000000100)
- Network 1 (00000000-0000-0000-0000-000000000110)
- Agent 1 (00000000-0000-0000-0000-000000000111)
- Agent 2 (00000000-0000-0000-0000-000000000112)
- Network 2 (00000000-0000-0000-0000-000000000120) ...
- Network n (...)
- Channel 2 (00000000-0000-0000-0000-000000000200) ...
- Channel n (...)
Now, let's assume we have these contents:
- Page A, tagged with "Channel 1"
- Page B, tagged with "Network 1"
- Page C, tagged With "Agent 1"
For example, if i am "Agent 1" and i want to obtain contents in this order
- Page C
- Page B
- Page A
(((owstaxIdTargeting:"GP0|#00000000-0000-0000-0000-000000000100" XRANK(cb=1))
owstaxIdTargeting:"GP0|#00000000-0000-0000-0000-000000000110" XRANK(cb=10))
owstaxIdTargeting:"GP0|#00000000-0000-0000-0000-000000000111" XRANK(cb=100))
Basically, i'm boosting contents created for "Agent 1", then contents for "Network 1", then "Channel 1".
Great? No. This method apparently works.
In this query i used XRANK, which boost the rank score... boost means SharePoint assign a score using ranking models you can boost manually using XRANK query directive.
This also means that rank scores could be scrambled by a lot of rank model rules, take a look to this articles:
- Overview of search result ranking in SharePoint Server 2013
- Customizing ranking models to improve relevance in SharePoint 2013
This model basically assigns a 0 score to all contents and it simply does NOTHING, so only XRANK values will be considered.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 | < rankingmodel2stage description = "A model with does nothing" id = "05E6223C-C4C0-4EB7-8E28-8242E48D05A9" name = "Fake Rank Model" xmlns = "urn:Microsoft.Search.Ranking.Model.2NN" > < rankingmodel2nn id = "242B7953-A3EE-499B-B6F5-A473AA33ADFA" > < hiddennodes count = "1" > < thresholds > < threshold >0</ threshold > </ thresholds > < layer2weights > < weight >0</ weight > </ layer2weights > </ hiddennodes > < rankingfeatures > </ rankingfeatures > </ rankingmodel2nn > </ rankingmodel2stage > |
0 commenti:
Post a Comment
Because of a lot of SPAM about courses, I need to moderate all comments here.
I ensure you that I will answer whenever possible (if you are not a spammer).