Data Mining Book?

HI, I am new to Data Mining. Would you please recommend any books on this area for starters? I'll appreciate your help greatly.
Hui

There is also another one at a basic level by Ian Witten and Eibe Frank.
Regards,
Scott Rappoport

Similar Messages

Help with data mining add ins-excel 2010

I've wasted hours of my life now trying to figure out how to establish a connection on the data mining add in on excel 2010. I have installed and re-installed the microsoft sql server 2012 express multiple times and don't understand what it takes
to get this to work...Do I need SQL server and why? Do I need to download the adventureworks data file and why? (For some reason I was able to download it properly on my first sql server install, but when I went to work the data mining connection it
said the SQL browser must be connected...What???...I thought it was connected...there are no instructions on how that is fixed...Now I have reinstalled sql 2012 but can't download adventureworks...it says it can't
establish a connection...I am beyond the point of frustration)...I'm not a developer or know anything about code/programming, so a lot of the lingo is way over my head anyway when I am searching for troubleshooting solutions...I just want to be able to use
this feature in excel and it is upsetting me that I can't get it to work. I have followed step by step instructions, watched youtube videos, etc...nothing doing...If anyone can help me it would be greatly appreciated. Thanks.

>I'm not a developer or know anything about code/programming,
You have two choices:
1. Hire a programmer type to assist you
2. Become a programmer
BOL: Data Mining Add-ins
Instead of Express, consider purchasing SQL Server 2012 Developer Edition:
http://www.amazon.com/SQL-Server-Developer-Edition-2012/dp/B007RFXQAM/ref=sr_1_1?s=software&ie=UTF8&qid=1397437432&sr=1-1&keywords=sql+server+2012+developer+edition
Install it.
Download AdventureWorks2012 and AdventureWorksDW2012 sample databases and install them.
Desirable (but not for beginner): install Adventure Works Cube. This is how it looks after installation:
Kalman Toth Database & OLAP Architect
Free T-SQL Scripts
New Book / Kindle: Exam 70-461 Bootcamp: Querying Microsoft SQL Server 2012

Something new in Data Mining in BI 7.0

Hello experts!
Does someone know if there is something new in Data Mining in BI 7.0? As far as I understand there is no more WebItem for prediction (Clustering and Decision Tree) and this is now integrated in SAP Analytics/Visual Composer.
Are there any new methods or functions. I can see nothing...
Best regards,
Peter

Hello all!
It is very sorrowful that no one seems to be interested in SAP BI Data Mining. There is even a new book from SAP Press about this topic.
So after all I can say that I had found nothing new so far. There are some new things in APD which are very interesting, but nothing new in Data Mining.
Peter

How to design a churn Data Mining application

Dear All,
I am a beginning software engineer. I am interested in designing a churn data mining applications for telecom companies.
My questions are:
1. Can I use the ODM sample codes to do this for creating the models,lifts, and scoring, etc?
2. How do I attach Java Swing front-end applications to the data mining in the Oracle 9i database server ?
3. Are the scoring posted in the sample codes accurate and useful in life deployment?
4. On the data warehouse for the data mining, what are the special considerations for the warehouse vis-vis dimensions, fact tables, etc?
Regards,
Olatuja Abayomi

Thanks :)
However, what i mean is the implementation (java file).
For example, the Java RMI tutorial at http://java.sun.com/docs/books/tutorial/rmi/overview.html defines two interfaces on server side.
One interface is for the Remote Interface, which has to be on server side.
Another is an interface with generic type, which is
public interface Task<T> {
T execute();
} Then both copiled interfaces (two class files) are compressed into a jar file, and this jar file needs to be reachable by both client and server side of the code.
So, client side needs the implementation of interface Task.
Or after we defined the two interface (Remote interface and Task interface) on server side, we also implement the Task interface
on server side, then compile the server side and compress compiled interfaces and implementation (three cless files) into a jar file, and pass this
jar file to client, so server needs no implementation for the interface Task, since the implementation is defined and compressed in the jar file already.
I hope i explained better this time, or i am just lost at this point.

Data warehousing book - guidance

can someone point out a good book on data warehousing with 10g, a book that focuses on concepts and details of the technology - more than focusing on how to use oracle tools- beginner/intermediate level?
thanks and sorry if the post is in the wrong context, i just thought i'd get guidance from people who use the technology

I would recommend looking at the following database documentation (these links are for 11g documentation but the majority of concepts are applicable to all versions of the database):
2 Day + Data Warehousing Guide
http://www.oracle.com/pls/db111/to_toc?pathname=server.111/b28314/toc.htm
Data Warehousing Guide
http://www.oracle.com/pls/db111/to_toc?pathname=server.111/b28313/toc.htm
OLAP User's Guide
http://www.oracle.com/pls/db111/to_toc?pathname=olap.111/b28124/toc.htm
Data Mining Concepts
http://www.oracle.com/pls/db111/to_toc?pathname=datamine.111/b28129/toc.htm
Also look at the whitepapers published on the Data Warehouse Technology Center home page on OTN : http://www.oracle.com/technology/tech/bi/index.html
Hope this helps
Keith Laker
Oracle EMEA Consulting
BI Blog: http://oraclebi.blogspot.com/
DM Blog: http://oracledmt.blogspot.com/
BI on Oracle: http://www.oracle.com/bi/
BI on OTN: http://www.oracle.com/technology/products/bi/
BI Samples: http://www.oracle.com/technology/products/bi/samples/

How to learn data mining?

Hi
I am an Oracle developer with DBA experience
How can I learn data mining and launch a career as a data miner?

Hi,
You can certainly start by using ODMr and walking through the tutorial.
To be expert in the field, I would suggest going to any of the book sites and searching for books on datmining. You will find alot of material useful in learning about the art of datamining.
Typically it is not just knowledge of datamining techniques that is important, but understanding the domain. If you have experience with domains such as retail or financial, then you should consider studying datamining implementations that target those domains. The more direct benefit you can bring your users the better.
Good luck, Mark

OLAP & Data Mining

Soon we are going into OLAP and Data Mining. For data mining Oracle provides Darwin tool. What are the tool it provides for OLAP. Since OLAP is a first stage for Data Mining. Please also suggest where I get these tools and books or demo.
Thanks & Regards
Shailesh

Hello. OLAP is not necessary for data mining, but it is complementary to it. For OLAP products, I suggest you check out these sites on OTN:
http://otn.oracle.com/products/exp_server/ and http://otn.oracle.com/products/express/
More information on Express products is also on www.oracle.com, at
http://www.oracle.com/ip/analyze/warehouse/bus_intell/
null

Problem solved by data mining??

what are the different problems that data mining can solve??

See book, here is the summary:
Data mining is concerned with discovering knowledge. It is about uncovering relationships or
patterns hidden in data that can be used to predict behaviors, outcomes, or provide some
other useful function.
The discoveries of data mining are typically cast into one of the following frameworks:
Classification—We have a customer, are they a high credit risk or low risk?
Clustering—We have a large universe of customers, can we define a small set of
customer types that does a good job, in some sense, of describing the entire universe?
Associations—When this happens, that tends to happen.
Scoring—What is the probability that this customer will cancel their service? What
is the likelihood that this is a fraudulent transaction?
In part, data mining is an application of the techniques of statistics. However, it frequently
goes outside of that field, employing methods of artificial intelligence or other mathematical
approaches to arrive at solutions that are more robust for working with messy, real‑world data

Data Mining on data specified and filtered by the user in runtime

Hi Experts,
i am new to Data Mining in SAP BI (we are on BI 7.0 SP Level 20). I familiarised myself with APD and Data Mining by reading some interesting and useful threads in this forum and some other resources. Therefore I got a understanding about the topic and was able to create basic data mining model for an association analysis and an corresponding APD for it and write the results into a DSO by using the data source. But for now I was not able to find a solution for a concrete customer requirement.
The user shall be able to select an article, a retail location and a month and get the top n combinations sold with that article in the particular location and month. For that he may not access the data mining workbench or any other SAP internal tools but he shall be able to start the analysis out of the portal (preferable a query).
We had some thoughts on the scenario. The first idea would be to create an APD for every location for the last month. As we need to cover more than 100 locations, this would not be practicable. Therefore I think it would be necessary, that the user can select the particular filters, and the data mining would then be executed with the given input.
The other idea was to use a query as source. The user would start this query and filter location and month in it. The result of the query could then be used as the source for the APD with the association analysis. Therefore we would need to create a jump point from that query, which starts the APD with that results. After that the user should be able to start a result query, which displays the result of the association analysis (ideally this result query would start automatically, but starting it manually would be ok, too).
So, I have the following questions for these scenarios:
1.) Is it possible to create variants of a single APD, for automatically doing the data mining for the different locations?
2.) is it possible to start an APD out of a query, with the particular results regarding filtering?
3.) Can we place a query directly on the data mining results (how?) or do we need to write the data mining results in a DSO first?
4.) What about the performance? Would it be practicable to do the data mining in runtime with the user waiting?
5.) Is the idea realistic at all? Do you have any other idea how to accomplish the requirement (e.g. without APD but with a query, specific filter and conditions)?
Edited by: Markus Maier on Jul 27, 2009 1:57 PM

Hi ,
you can see the example : go to se 80 then select BSP Application ,SBSPEXT_HTMLB   then select tableview.bsp , you will get some idea to be more clear for the code which you have written
DATA: tv TYPE REF TO CL_HTMLB_TABLEVIEW.
tv ?= cl_htmlb_manager=>get_data(
                         request = runtime->server->request
                          name    = 'tableView'
                              id      = ''tbl_o_table" ).
IF tv IS NOT INITIAL.
DATA: tv_data TYPE REF TO CL_HTMLB_EVENT_TABLEVIEW.
tv_data = tv->data.
IF tv_data->prevSelectedRowIndex IS NOT INITIAL.
FIELD-SYMBOLS: <row> LIKE LINE OF sflight.
    READ TABLE ur tablename INDEX tv_data->prevSelectedRowIndex ASSIGNING <row>.
    DATA value TYPE STRING.
    value = tv_data->GET_CELL_ID( row_index    =
                               tv_data->prevSelectedRowIndex
                                  column_index = '1' ).
endif.
endif,

BW Data Mining

Hello!
I need information about the following BW Data Mining topics. Hope, someone can help me?
-What 3rd Party Tools are available für Data Mining? (I just know "IBM Intelligent Miner")
-Does someone have experiences with performance of the DM-algorithms? System requirements? Mass data?
-Are there possibilities to improve the performance?
-What is known about new features in Netweaver BI 7.0 for DM?
-Does someone have special experiences in DM in the automotive industy? Are there typical business processes known, which are useful for BW Data Mining?
Thank you for help!
Regards,
Peter

Hallo!
I need different information about BW data mining.
One thing is, what are typical business scenarios for the different algorithms. So it is easy to find such for decision trees, association analysis and maybe for ABC analysis. But i have little problems to find real scenarios vor weighted score tables and regression analysis. It would be helpful to hear about some.
I have to write a paper for my company about data mining with bw. While we have no experiences in this area I´m looking here for people having.
Also Performance. The problem is not to try out with APD and Data Mining models. But to find experiences with this.
Good or bad performance with this or this algo? Or how can I optimize performance for mass data?
After this I have to design typical business scenarios. Especially for automotive area if possible.
So what the question for me is, what is possible with the algorithms? Where are problems with them?
It would be great if you could provide me with more information about your experiences in this area.
Thanks,
Peter

Data Mining overflow error while loading the mining viewer

I developed a time series model that processes succesfully. The problem I have is that when I try to view the mining model viewer that displays the times series on a chart, I get the error below:
An error occurred while a prediction query was being executed:
'Error (Data mining): An overflow was encountered while converting a predicted value to the '' column, at line 1, column 33. A higher precision data type for the column is recommended.'.
ADDITIONAL INFORMATION:
Error (Data mining): An overflow was encountered while converting a predicted value to the '' column, at line 1, column 33. A higher precision data type for the column is recommended. (Microsoft OLE DB Provider for Analysis Services 2008 R2.)
It looks like the probability value (which is a float data type, for example 0.4452938765) is being stored in a particular column, and the data type of that column cannot contain the range of decimal places that the probability has.
I don't know which column to look for to change its data type like the error messages says. Would anyone have any idea?

Hello,
Thanks for your posting.
It is more related to Data Mining issue in this case. I move the thread to
Data Mining forum for better support. Thanks for your understanding.
Regards,
Elvis Long
TechNet Community Support

Data mining is Loading after upgrade from 10.1.0.4 to 10.2.0.4

SQL> select comp_name, version, status from dba_registry;
COMP_NAME
VERSION STATUS
Oracle Ultra Search
10.1.0.4.0 NO SCRIPT
Oracle XML Database
10.2.0.4.0 VALID
Oracle Enterprise Manager
10.2.0.4.0 VALID
COMP_NAME
VERSION STATUS
Oracle Text
10.2.0.4.0 VALID
Oracle interMedia
10.2.0.4.0 VALID
Oracle Expression Filter
10.2.0.4.0 VALID
COMP_NAME
VERSION STATUS
Oracle Workspace Manager
10.2.0.4.3 VALID
Oracle Data Mining
LOADING
Oracle Database Catalog Views
10.2.0.4.0 VALID
COMP_NAME
VERSION STATUS
Oracle Database Packages and Types
10.2.0.4.0 VALID
JServer JAVA Virtual Machine
10.2.0.4.0 VALID
Oracle XDK
10.2.0.4.0 VALID
** How to fix ti..
*** I can find below error in dbua log
===
dbua
Oracle_Server.log 470085 select dbms_java.full_ncomp_enabled from dual;
470086 select dbms_java.full_ncomp_enabled from dual
470087 *
470088 ERROR at line 1:
470089 ORA-29558: JAccelerator (NCOMP) not installed. Refer to Install Guide for
470090 instructions.
470091 ORA-06512: at "SYS.DBMS_JAVA", line 236
470092
470093
470094 Rem If Intermedia, Ultrasearch, Spatial, Data Mining upgrade,
470095 Rem first install JAVAVM if it is not loaded
470096
470097 BEGIN
470098 2 IF dbms_registry.is_loaded('JAVAVM') IS NULL AND
470099 3 (dbms_registry.is_loaded('ORDIM') IS NOT NULL OR
470100 4 dbms_registry.is_loaded('WK') IS NOT NULL OR
470101 5 dbms_registry.is_loaded('SDO') IS NOT NULL OR
470102 6 dbms_registry.is_loaded('EXF') IS NOT NULL OR
470103 7 dbms_registry.is_loaded('ODM') IS NOT NULL) THEN
470104 8 :dbinst_name := dbms_registry_server.JAVAVM_path || 'initjvm.sql';
470105 9 ELSE
470106 10 :dbinst_name := dbms_registry.nothing_script;
470107 11 END IF;
470108 12 END;
470109 13 /
470110
470111 PL/SQL procedure successfully completed.
470112
470113 SELECT :dbinst_name FROM DUAL;
470114
*** Then I can apply 10.2.0.5 PSR ( in data mining is loading)?

You can try de-install and install this component using the following note
Master Note for Oracle Data Mining (Doc ID 1087643.1)

Error (Data mining): The specified mining structure does not contain a valid model for the current task.

I'm trying to run the Cross validation report on a mining structure that contains just Microsoft Association Rules mining model. In Target Attribute, I've tried:
Actual(Service Description).SE value
Actual([Service Description]).[SE value]
Actual(Service Description)
Actual([Service Description])
just because i don't know what is the exact correct format, but none of them worked, and I always get the following error:
Error (Data mining): The specified mining structure does not contain a valid model for the current task.
the following is my mining model structure

Association rules does not allow for cross-validation
Mark Tabladillo PhD (MVP, SAS Expert; MCT, MCITP, MCAD .NET) http://www.marktab.net

Data mining Algorithms in Essbase

Hi,
Just wondering if anyone has used data mining algorithms provided within Essbase. Any thoughts or pointers towards more information will be helpful..
Thanks in Advance !!

In a 2009 persentation at Kscope from ODTUG titled little used features of Essbase, I went through how to use data moning. It is available on the odtug website. I do know that nothing has been done with the data mining modules in a long time as the team was disbanded since Oracle has other tools to do data mining.

What are the options for me for Data Mining tools?

I am new for data mining. I have a project for which we are asked to identify the relationships between variables we got and we were asked to come up with an algorithm from data mining. What are the options available for me as data mining tools and where
do I start from? How do I travel through this process. Please advice.
mayooran99

I think that Predixion Insight is the easiest tool to use and is very powerful.
All you need to have is Windows OS, Office 2010 or 2013 and internet connection. If you want to look at what you need to do before installing Predixion, I recommend you to take a look at the walkthroughs at
http://predixionsoftware.com/Help/webframe.html#Data%20Modeling%20Walkthroughs.html
To install Predixion, go to
http://predixionsoftware.com/Technology/Predixion-Insight-Download
Tatyana Yakushev [PredixionSoftware.com]

Data Mining Book?

Similar Messages

Maybe you are looking for