Frequently Asked Questions about TextualETL™
Click on a question located below to reveal the answer.
- Can TextualETL™ operate on any electronic source of data?
Either directly or indirectly, yes.
- What base language is TextualETL™ written in?
- Can TextualETL™ produce output for any relational data base?
Yes. This includes Oracle, SQL Server, DB2, Teradata, Netezza, etc.
- What types of documents can TextualETL™ handle?
Highly repetitive, moderately repetitive, and non repetitive documents; Textual ETL handles ANY structuring of documents.
- How many different analytical functions does TextualETL™ comprise to handle these different document structure types?
At last count – 42 and growing.
- Can different analytical functions be used all in the same pass of text all at the same time?
- Will passing text through TextualETL™ increase or decrease my total volume of data?
It depends on the document and the type of processing that is done. It widely varies, from case to case.
- What are some of the typical dbms that are supported by TextualETL™?
SQL Server, Oracle, DB2, Teradata, and more
- Can I run TextualETL™ in a Teradata only environment?
- How many different types of data bases are produced by TextualETL™?
Depending on the document and what you want to have done, up to seven types of data bases can be created
- Do I have to do data base design to use TextualETL™?
No. The data base design is already done
- Are stop word lists shipped with the product?
Yes, in multiple languages
- What languages are supported?
English, Spanish, Portuguese, French, German, and Italian currently.
- Are taxonomies available with TextualETL™?
Yes, through our partner Wand, Inc.
- If I have already built my own taxonomy, can I use it?
- How much taxonomy is available?
Approximately 49,000 taxonomies spread over 4.3 million words in 16 languages, covering every subject imaginable
- Can TextualETL™ read and support email?
- Can TextualETL™ support call centers?
- Can TextualETL™ support Hadoop and MapReduce?
- Does TextualETL™ directly support HTML?
- Can the taxonomies be customized if purchased by Wand?
- Can the taxonomies be managed independently of TextualETL™?
- Is TextualETL™ scalable?
- Can TextualETL™ support tweets?
- What are some of the typical applications that are used in conjunction with TextualETL™?
Contracts analysis, government filings analysis, email analysis, warranty claims analysis, medical records analysis, and so many, many more.
- Can TextualETL™ be used in a SaaS environment?
- How is TextualETL™ purchased?
Typically by the server CPU with an unlimited number of users or as a cloud-based SaaS (“software-as-a-service”) solution.
- Does TextualETL™ have consulting services that accompany the product?
Forest Rim will help make the arrangements for consulting
- When I create a data base with TextualETL™, can I use that data base in conjunction with analysis of classical structured data?
- Does TextualETL™ support OCR?
Yes. TextualETL™ takes the output of OCR and uses that as input into TextualETL™. TextualETL™ is not a replacement for OCR
- Once the mapping is done for a document type in TextualETL™, can I take the mapping and save it and use it again on another day?
- If I have done a mapping and I am processing a very similar document type, can I make use of the previously created mapping?
- Once I have created my output databases, what Business Intelligence tools can I use to analyze it?
You can use any tool that you have already been using. The analyst doing business intelligence only knows that there is a new source of data. But nothing else changes
- Does Forest Rim offer other supporting analytical tools?
Yes. Occasionally you need to look at data in the context of its own text. TextualETL™ supports that form of analytical processing with a tool we call Textual Business Intelligence.
- Does TextualETL™ modify the originating source document?
- Can I go from the data bases produced by TextualETL™ back to the originating source document?
- How long does it take to install TextualETL™?
Usually about 30 minutes for standalone single-CPU solutions. Our cloud-based SAS solution is similar. Enterprise-wide solutions can be more intensive and time-consuming when integration with multiple data-sources is required.
- From the time I start with TextualETL™ until I start to see the first of my results, how long is that period of time?
You meet the consultant at 9:00 am, you install TextualETL™ by 10:00 am, and you can be using the first of your results by 3:00 pm in the afternoon, all in the same day.
- What prerequisite software is there for TextualETL™?
If you are running in a Microsoft environment, you need Office and SQl Server 2008 or later, as well as the vb.net framework. A minimum of 2 CPU’s (dual-core) is suggested and quad core processor is recommended.
- Can Forest Rim do a proof of concept?
Yes. Forest Rim does free Proofs of Concept all the time. We typically run the proof of concept on our environment as scheduling permits. POC’s performed on premise my add additional cost and expense for time, materials, and travel. Please contact us for details.
- Does Forest Rim offer ongoing support and maintenance?
- Is Forest Rim used for bulk loads of text?
In most circumstances, Forest Rim is used for iterative running of documents, not for bulk load processing.
- Can Forest Rim take text from spreadsheets?
Yes,there are some limitations in using spreadsheets as a source. However, these limitations can be overcome or circumvented with additional customization or programming using our off-the-shelf API’s.
- Is there documentation to support TextualETL™?
- Is TextualETL™ for the technician or the end user analyst?
The end user analyst.
- Who is Forest Rim’s competition?
There certainly are other companies that manage unstructured data. But all of the companies that we are aware of do not do what Forest Rim does. Autonomy is good for defining taxonomies and looking at relationships in the world of text. Endeca is good for doing the analytics from text. And other companies have their niches. But Forest Rim is the only company that we know of the can read any form of text in any structure of text and interpret that text and create any relational data base management system output that you want.
- How does TextualETL™ compare to Map Reduce?
Map Reduce is peculiar to the Big Data environment and is designed to handle huge volumes of data. It is written for the technician to navigate through text. And Map Reduce operates on non repetitive data. On the other hand TextualETL™ is designed to handle non repetitive data and repetitive data (as well as mildly repetitive data.) TextualETL™ is designed for the end user business analyst, not the technician. TextualETL™ has access to 49,000 taxonomies that can be used to categorize text. TextualETL™ is sensitive to the logical structure of text. No one confuses a chain saw and a scalpel. If you want to become a logger, you need a chain saw, not a scalpel. And if you want to do open heart surgery you need a scalpel, not a chain saw. Map Reduce is a chain saw, TextualETL™ is a scalpel.
- Can you do data quality processing inside the TextualETL™ engine?
Only in the crudest sense can you do data quality inside TextualETL™. TextualETL™ is designed for accessing and interpreting text – putting context on text. Once you pull the text out of TextualETL™ you are free to do all sorts of data quality analysis. TextualETL™ is designed to do the very difficult job of reading and interpreting raw text – any raw text. After text has been run through TextualETL™, the context of the text is understood and the meaning of the text is recognized. Do the recognition and interpretation of text is a very complex task. So TextualETL™ sets the stage for data quality but in reality, TextualETL™ is not a data quality tool.
- How can I try TextualETL™?
Forest Rim can provide a demo of our software or perform a proof of concept for you. Please contact us today to schedule your demo or POC discussion.
- If the client would prefer to limit iterations, getting 90% of 3 fields in one pass is better than getting 12% of all fields after 6 passes. How do you recommend the client approach deciding on how big/small a net should be used?
Unfortunately iterative development of the parameters needed to make TextualETL™ is mandatory. That is due to the irregular nature of text. So doing away with iterative development is not a possibility. However, after enough iterations have been run, there develops a base set of parameters. Once the base set of parameters have been established, then an organization can run those parameters against a large set of documents and expect to have very few documents that are not in conformance. But in any case the initial creation of the parameters is a necessary step. That is one of the reasons why Forest Rim insists on having consulting partners being sold along with the product.
- Once I have created my indexes/data bases after doing textual ETL processing, can I edit my indexes before putting them in production?
Yes. Forest Rim supports a robust set of edits which can be made against the indexes/data bases produced by TextualETL™.
- Does textual ETL do analytical processing?
No. TextualETL™ sets the stage for analytical processing. Once TextualETL™ has “done its thing”, then analytic processing packages are able to be placed on top of the results of TextualETL™. But in any case, TextualETL™ is an enabler of analytic processing, not an analytic processing engine itself.
- Can TextualETL™ be used to classify documents into broad categories?
- Once Forest Rim has created its tables/output, can you link that data to standard operational data?
Yes. Forest Rim embeds metadata inside its tables that can be used for linkage between the structured and unstructured world. It is actually easy to do.