Search Relevance %

Hi,
In our defined KM Search, the results are sorted by relevance. But the relevance % doesn't go with the matches displayed in the document abstract. A high relevance % has 2 matches vs a lower relevance has 4 matches.
Does anyone know how the following:
1) How is this relevance % is calculated?
2) Is it solely related to the indexing done by TREX?
Any help would be much appreciated. PD We have Trex 6.0.
Regards

The TREX text-mining engine enhances the functions of TREX by reducing document content to meaningful words with the help of text operations such as reduction to root form, normalization of spelling, the removal of stop words, and so on. This improves retrieval results. The search results are refined because the relevance of the documents can be evaluated differentially. This is possible because the similarity between the documents is calculated on the basis of the tried and tested vector space model: Each document and term in the document set is represented by a multi-dimensional space. This representation allows TREX to carry out enhanced retrieval and classification methods. In this way, documents can be ordered sensibly according to certain criteria by being assigned automatically to a generic category or class (document classification). When a document is found, the user also receives a list of key words that give an overview of document content and help to characterize the document and distinguish it from others (determination of key words). The user can also look for documents that are similar to the one found. This helps to restrict the number of documents that may be relevant (Search for similar documents). TREX offers terms that are similar to the original search term in content and meaning. The user can then reformulate the search request (Search for similar terms).
Have a look at this [PDF Link|https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/81c1ca90-0201-0010-a2a1-b0412ec9cd1a]
Regards
BP

Similar Messages

Knowledge Search Relevancy

Hello,
The knowledge search results on webclient have a relevancy for each result.
Can anyone explain
1. How this factor is calculated?
2. How to influence this field if we think that this is not accurate?
3. Does the search includes dependent knowledge base in the search results? For example, if we search solution database (SDB), is the SDB attachment knowledge base that is configured as a dependent knowledge base of SDB considered in the search?
I know that SAF does something to assign a relevancy factor. But need to know more details regarding this.
thanks
Prakash

Hi Prakash,
I also need information about the Knowledge search relevancy and how this is calculated. Did you manage to find anything out about this?
Kind Regards
JoJo

BADI for Claims search in trade promotion management

Hi Experts,
I am enhancing the Claims search in trade promotion management with a custom field. I added a custom field in the search screen of claims.
I want to filter the records based on that custom field. The component is BT311S_CSR and the search query is BTQCSR.
Is there any BADI for claims so that I can filter the records.
I tried to put a breakpoint in the badi CRM_BADI_RF_Q1O_SEARCH and searched but it is not stopping.
Does this badi works for claims? Or first do i have to create a implementation with the filter object as BTQCSR then try to check whether it will stop in the badi while searching the claims.
Please suggest me.
Regards,
Dinesh.

Hi Dinesh,
Now this becomes tricky. Since your requirement is very specific there is no way that standard sap can help. You may end up in enhancing n number of classes.
I can propose a solution though Please put of in front of your business/ functional team and then go ahead..
Now we will change our approach...Please follow below steps:
1. Add a altogether new field to your claim transaction using AET for sales office (description as 'Sales Office').
2. Make this field search and result relevant.
3. Whenever a sold to party is selected for claim transaction, based on the sold to party you would set value of this field.
4. Since your field is search relevant it will automatically appear in your search query (control this in view configuration.)
5. You may or may not display this field to end user in claim detail screen, You can control this by configuration.
6. Depending on your requirement you can show/hide standard sales office field in search query.
If you follow this approach then it will serve the purpose.
Trust this solves your issue. Give points if it helps.
Regards,
Bhushan

Operation of Search architecture.

Good morning MS community,
So far i have been done the SharePoint Server 2013 search architecture:
http://technet.microsoft.com/en-us/library/cc263199%28v=office.15%29.aspx
Afterward, i myself interpret the diagram into the image below: This is an operation of SharePoint 2013 search from my understanding.
My question is: "Could you please advise me whether i understand correctly about the "search architecture" - SharePoint 2013 - from my below analytic?"
******************* My analysis of "Search operation " *****************************
1. User enter a string of information that he/she wants to looking for. The string is sent to query components.
2. Query processing components will:
- Analyze the Search queries.
- Perform linguistic processing. Such as "word-breaking" and "stemming".
- Optimize precision, recall, relevancy.
- Submit to index component.
* Later on, index component will gather data from 2 source:
- Crawl and components process.
- Analytic process.
3. Gathering Data from Crawl and components.
3.1.
Crawl component will collect crawling content sources: invoke connectors/ protocol handler to interact with content source to retrieve data.
- Type of data:
+ Actual data.
+ Metadata.
- At the same time, the crawl components will:
+ Store information about crawled items on crawled DB. Crawl component will write the following information to crawl Database:
* Last crawl time.
* Last crawl ID.
* Type of updating during last crawl.
+ Track crawl history.
è After successfully gathering information from content resource, the crawled
data will be sent to "content processing component".
3.2. Content processing component:
- Transform crawled items into "artifacts" that can be included in "search index".
+ How to transform: performing some operation such as "document parsing
and property mapping".
- Perform linguistic process, such as "language detection + entity extraction".
- At the same time, content processing component will write information about "links + URL" to the links database.
è When finish, content processing will send data to index component.
4. Gathering data from Analytic process.
* Explain about components in this part:
4.1. Analytic process component.
- Operations: search analytic + usage analytic.
- Aims: improve search relevance, create search reports, and generate recommendations and deep links.
4.1.1. Search analytic:
- Extracting information
+ Link
+ Number of times clicked.
+ Anchor text.
+ Data related to people.
+ Metadata.
4.1.2. Usage analytic:
- Analyze usage log information. (retrieve this info from the font-end via event store).
è Generate usage and statistic reports, store in "analytics reporting
Database".
4.2. Analytic reporting database.
- Result of usage analytics: usage + statistic reports.
4.3. Link database.
- Store information of:
+ search clicks.
+ Number of times people click on a search result from the search result page.
4.4. Event store:
- Usage events ( such as number of times viewed).
*** operation of "analytic processing component ***
- Get information from "link DB", "analytic DB", "event store".
- Extracting information (links, number of times an items clicked, anchor text, data related to people, metadata …)
- Send information to "index components".
- At the same time, write new information to "link DB" + "analytic DB".
5. "index and query process"
- Receive processed items from "content processing components" "analytic processing components"
è write those items to an index file.
- Send back results to "query processing component".
6. "Query processing component" Return a set (results) based on the processed query back to the query processing component.

Hi,
Search in SharePoint 2013 has been re-architected. You could refer Figure 1 from the link below for big picture of Search process and components:
http://msdn.microsoft.com/en-us/library/office/jj163300(v=office.15).aspx
For some basic concepts, you could refer to:
http://technet.microsoft.com/en-us/library/jj219738(v=office.15).aspx
If you would like more detail information about search, you might need to contact Microsoft support engineer for more sufficient information.
Regards,
Rebecca Tu
TechNet Community Support

Search across multiple site collections SharePoint 2013

Hi Guys
I have a search center and Enterprise search configured in one site collection, I need to enable it to search content in other site collections too. I am using SharePoint Server 2013 Enterprise on a Windows Server 2012 R2. Can I promote results from
a particular site collection if the user initiates search from that particular site collection..?
Thanks in Advance
Parth

Hi,
From your description, you want to find a way to promote result from a particular site collection when users search on
Enterprise Search site.
To add promoted result, you can create new query rule and add
Promoted Result or Result Block to highlight the results you want.
Moreover, What do you mean when you say promote results from a particular site collection if the user initiates search from that particular site collection?
A real sample? Please explain more about your requirement so that we can provide more detailed information.
If you are new to SharePoint 2013 search, you may start from:
Search in SharePoint Server 2013
http://technet.microsoft.com/en-us/sharepoint/jj898538.aspx
Manage
Manage the      search schema
Manage the Search Center
Manage search      relevance
Manage crawling
Manage the      search topology
Back up      and
restore
Use      Windows PowerShell cmdlets to manage search
Miles LI TechNet Community Support

ARQ: Does LDAP User Search action require any special authorization for requester???

Hi All,
I was wondering if requester need to be given any special authorization to search users in LDAP?
Because, I have noticed that a requester can not search users from LDAP. However, another user who is a super user in GRC system and has SAP_ALL profile assigned, can search users from LDAP easily!
I have noticed only this change between these two users and not sure what authorizations should be granted to requester to search users from LDAP. I have tried to search relevant auth. object in his role "SAP_GRAC_ACCESS_REQUESTER" but could not find. I also check security guide for this but did not get any details.
Can anyone advise?
Regards,
Faisal

Alessandro,
I switched on trace using ST01 for one of the requesters and viewed its details later. I found RC=4 or 12 for some of the auth. objects.
For example:
I opened of of the records and could see above details. I am unable to interpret it further. Can you please assist in this?
Regards,
Faisal

Relevance of title field in doc library

Bit of a debate at work and I'm trying to dig up some hard facts.
We have been told that the title field in a document library and is pointless and can lead to user confusion between title/filename and basically adds no value. There has been recent talk of hiding the title field altogether from forms.
My concern is the effect this may have on search results.
We have our own custom meta data fields so this question doesn't relate to collecting meta data - we are covered there.
I have hunted down some info and have read comments saying that title is second only to content when it comes to the search algorithm. I've also read other articles that are less specific and just say it is good.
Is there a definitive answer?
This is a 2010 question but will also apply for our upgrade to 2013 which will be happening soon.
1. How does the title field feature when ranking results.
2. Is there a list/URL that defines all aspects of relevance building?

Hi ,
Relevance is about how closely the search results that are returned to the user match what the user wanted to find.
The formula to calculate relevance uses two areas of ranking called Static and Dynamic ranking.
Dynamic Ranking (query-dependent ranking) is where the property values or content of an item affects the ranking score. As example the ‘Title’ field can be evaluated against the search criteria and the more important we consider the field
to be, the higher the ranking score will be for items where the value in the ‘Title’ field have a closer match to the search criteria.
Static Ranking (query-independent ranking) is where the content or property values of an item do not determine the ranking of the item.
Dynamic Ranking contains Title Extraction - only performed on Microsoft Office files. In scenarios where the ‘Title’ field of an Office file does not accurately reflect the contents of the item (example when a title of a file is ‘Presentation
1’ or ‘Document 1’), Enterprise Search detects another candidate for title within the body of the content item, and includes this value with the actual title when calculating relevance.
More information, please refer to the link:
http://social.technet.microsoft.com/wiki/contents/articles/20649.sharepoint-2010-improve-search-relevance.aspx
I hope this helps.
Thanks,
Wendy
Forum Support
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact
[email protected]
Wendy Li
TechNet Community Support

Attachment knowledge search DB

Hello,
I would like to ask you about knowledge search in CIC (4.0). The question is that we want to migrate the data to SAP from another Data base and be able to search in the document attached. But it seems like LSMW does not download documents, therefore how could we make it?
Thanks in advance

Hi Prakash,
I also need information about the Knowledge search relevancy and how this is calculated. Did you manage to find anything out about this?
Kind Regards
JoJo

How good Endeca Search is for ECM (Search on documents) products

HI all,
I am evaluating Endeca Search (Guided Search) as a search engine for Enterprise content management's .
Basic objectives are -
1) Search engine should be able to search documents with high relevancy.
2) When user sees the search results , he /she should be able to navigate/open the original document.
To be precise, users need is to find and open documents.
So I expect different points of view for the above use case. If anybody implemented similar use case, then please make some noise .
Regards,
Hoque

So I have been involved in half a dozen enterprise search projects using Endeca. Things to consider:
1. Ingesting the documents. (velocity, variety, volume and frequency of change, location of files ). Endeca Content Acquisition Server comes with a variety of connectors for various industry standard document repositories, web and file systems crawls. Document converter for various document types is included.
2. ACL and profiles for users. It is easy enough to create a link to the doc repository and pass credentials of the user. This has been fairly easy to achieve in the past.
3. Search relevancy ( Endeca has a versatile relevancy ranking technology)
4. Apply facets to documents so that they can be refined. Various ways to achieve this.
5 I18N Are the documents one language, multi language.
Check your price points for doing this , there are some systems out there that index everything and provide a basic text search, Endeca can combine structured and unstructured records with faceted navigation but it needs effort and work to do this.
hth
Terry

Search Page to All SAP-related Sites

General point:
I have created a web page that searches relevent SAP-related sites on the net.
<a href="http://www.allthingssap.com">www.allthingssap.com</a>
It's a nice way to narrow your search from the start and not get a bunch of junk in the results. The sites it searches are listed on the entry page. If anyone would like another site added email me (from the contact us link on that page) and I'll put it on the list.
Ken Greenwood

Hi Ken,
Yes, it was the ABAP class in Sarnia! These days I am primarily a Workflow consultant and do not do much ABAPing. I would like to learn the new ABAP capabilities though! Drop me a line if you can. My email address can be found on <a href="https://www.sdn.sap.comhttp://www.sdn.sap.comhttp://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/com.sap.sdn.businesscard.sdnbusinesscard?u=ulbojyzfq%2b0%3d">my SDN business card.</a>
Cheers,
Ramki Maley.

Error while adding a custom field with Input help via AET

Hi All,
I need to add two custom field under Service orders at Item level in component BT140I_SRVP.
One field is required to have the input search help f4 and autopopulates the second field
I am able to add one field(not requiring help) successfully through AET .
I have created one Zsearch_help in se11 and its successfully running and Autopopulating seocnd field while I am testing it
While adding second field through AET,I need to enter following details as -
field label,search relevant ,serach help etc.
When I type the name of my 'Zsearch_help' against field search help it gives me following error
'Search help is not compatible'.
Secondly,not getting getter and setter methods for the attrributes in BTAdminI.
Last,please tell me if i create zhelp and activate it,would it automatically appear in the list on AETwhile assiging it to input field?
Please help me out.Kindly be detailed as I am new to SAP CRM.
Thanks,
Shivani

The question is answered in CRM 7.0 forum:
Getting error while adding a custom field (with input help) through AET

Implementing KM scheduler in WD application

HI All,
Can anybody guide me on KM schedulers and how to implement the same in WD applications.
We have a WD application where we are creating PDF forms. Now we need to schedule the process and want to run the job at a regular interval of time. Is KM scheduler the right way to do it?
Please suggest on how to implement the same.
Regards,
Becky.

Hi Ashok,
Yes, u were correct. I didnt configured my Index at portal correctly and so was the issue.
Now I am able to get search-results.
Thank you.
Now can you tell me about what all parameters I could get from these search results?
I mean I tried Standard Search URL of portal :-
http://serverHost:serverPort/irj/servlet/prt/portal/prteventname/HtmlbEvent/prtroot/com.sap.km.cm.basicsearch?SearchType=seealso&SeeAlsoUri=
It shows various other details along with what we get using KM API.
for example: search relevance percentage, seached text etc
How do I get these parameters from API?
Thanks and regards,
Amey Mogare

Getting error while adding a custom field (with input help) through AET

Hi All,
I need to add two custom field under Service orders at Item level in component BT140I_SRVP.
One field is required to have the input search help f4 and autopopulates the second field
I am able to add one field(not requiring help) successfully through AET .
I have created one Zsearch_help in se11 and its successfully running while I am testing it
While adding fsecond field through AET,I need to enter following details as -
field label,search relevant ,serach help etc.
When I type the name of my 'Zsearch_help' against field search help it gives me following error
'Search help is not compatible'.
Please help me out.Kindly be detailed as I am new to SAP CRM.

Thanks for very helpful reply. After implementing the suggested SAP note, I am able to see the getter and setter methods.
So one of my problem has got solved with your kind help:)
As per your another suggestion,I have created the enhanced fields without the search help from AET in the node BTAdminI with names Plant(ZZPLANT) and Storage Location(ZZStoarge_Loc).
Now I am facing below problem:
Since Plant needs to autopopulated by Storage Location (search Help ZOFI_SHLP_STORAGE_LOC),So I need to Implement Getter setter method.
I have written below code into Get_V_ZZStorage_Loc Method:
method GET_V_ZZSTORAGE_LOC.
DATA:
    ls_map    TYPE if_bsp_wd_valuehelp_f4descr=>gtype_param_mapping,
    lt_inmap TYPE if_bsp_wd_valuehelp_f4descr=>gtype_param_mapping_tab,
    lt_outmap TYPE if_bsp_wd_valuehelp_f4descr=>gtype_param_mapping_tab.
    ls_map-context_attr = 'ZZSTORAGE_LOC'.
    ls_map-f4_attr      = 'STORAGE_LOC'. (Storage_loc is the parameter defined in search help)
    APPEND ls_map TO: lt_inmap, lt_outmap.
CREATE OBJECT rv_valuehelp_descriptor TYPE cl_bsp_wd_valuehelp_f4descr
    EXPORTING
      iv_help_id        = 'ZOFI_SHLP_STORAGE_LOC1'
      iv_help_id_kind   = if_bsp_wd_valuehelp_f4descr=>help_id_kind_callback
      iv_input_mapping = lt_inmap
      iv_output_mapping = lt_outmap.
endmethod.
But its not working.Should I write some code in Get_P_XX or some other method also?
Kindly suggest .

Re: CD/DVD drive does not seem to start DVDs in Satellite P750

My CD/DVD drive does not seem to start DVDs, it would turn and turn and the only way I can start a DVD (i.e. a movie) is by using Windows Media Centre, while software to be installed needs to be activated from Windows Explorer.
Anybody?
Please help.
Thanks!

Thanks, Xaver, tried this, still won't autoplay, it will just turn the dvd around for a while and then stop. Though when I open Windows Medi Centre and choos 'play DVD', it does the trick. The only way to install software, though si through Windows Exp[lorer and searching relevant file. Cheers,
Havoc.

External Service Procurement : Queries

I have three queries regarding external Service Procurement
1) IN SERVICES Purchase order , Over delivery Tolerance and unlimited TOLERANCE can be defined at two places ,
A) in Service tab aginst each service
B) iN po LINE ITEM , DELIVERY tAB
What is the use of defining at two places and how system works , if I define 0% Overdelivery in Services but checked "Unlimited" in Delivery tab , how system will behave ?
2) If I make OverDelivery tolerance in PO Line detail->delivery Tab as ZERO , UNLIMITED automatically gets changed to YES (Flaged) ... Why it is so ? I may have cases where I dont want any Overdeliver Tolerances ?
3) In SPRO -> MM-> EXTERNAL SERVICES Mgmt -> Define Screen layout ..
we can define our screen layouts for following combinations ( as per SAP Help)
1) Object (e.g. purchase order)
2) Document type (e.g. NB)
3) Transaction code (e.g. SAP transaction ME21N)
4) Release indicator (in the standard system: B)
You can define your own field selection strings for the specified release indicator. The field selection string must be identical with the relevant release indicator.
5) Item category (e.g. D for external service or B for blanket purchase order) e.g. ME21N, "Create purchase order".
What is the priority of searching relevant Screen layout ,
like If i have defined one field (Say service number) , mandatory at Object level , Display at Document level and optional at Tocode level ... Will system show me as Display/mandatory or optional ?
Pls do reply ...
Regards
Shrey

Hi DELHI INDIA,
1) when you are creating Service PO , you may enter service no. or just a short text . And in the item level service tab you enter the detailed service list . Now when you enter the overdelivery and unlimited details at
A) in Service tab aginst each service - it is only valid for that individual service
B) iN po LINE ITEM , DELIVERY tAB - it is valid for the whole line item
But always A overrides the conditions of B.
I mean If you maintain it at A level , then it will follow the rule of A instead of B
"if I define 0% Overdelivery in Services but checked "Unlimited" in Delivery tab " the system does not allow you overdelivery . But if you have any line in services where you do not set the overdelivery limit , then for that it will take unlimited overdelivery .
2) It may be because in the item level services you have checked unlimited ( by default it is unlimited) for all the services
3)
" have defined one field (Say service number) , mandatory at Object level " - so while creating that object you have to enter the field value .eg while creating PO , vendor is mandatory3
", Display at Document level " - while creating or changing the document it wil not allow it to chane the field value as it is at display level
"optional at Tocode level " - means while executing that Tcode that field is optional , eg in MMR gross weight of a material is optional . You may or may not enter the field value.
If it does not help please elaborate it .
"Will system show me as Display/mandatory or optional ?" -- I am not getting you exacty.
Hope it helps,
Regards,
Anupam

Search Relevance %

Similar Messages

Maybe you are looking for