Overview
Introduction
Service Features
Service Process
Output Quality
 |
 |
||||||||||||||||
|
INTRODUCTION TO AUTOMATED DOCUMENT PROCESSING Beginner What is automated document processing? ADP, sometimes called machine-assisted document processing, is the systematic use of technology to assist human beings in performing document processing tasks. Typical tasks in litigation support include:
What is unitization? Unitization, also called logical document determination, is a process that groups pages into their proper documents. The technique ensures that no pages are inadvertently mixed together or accidentally split apart at scan time. Unitization also results in the strongest document encoding because there is no omission or mixing of pertinent document information. Both electronic files and paper documents can be unitized. What is document encoding? Document encoding, also called indexing, is the process of creating a database of bibliographic information about a set of documents. Bibliographic information covers information about the document, such as its title, its creation date and its author. Historically, trained paralegals and legal temp workers performed this task by hand. Automated encoding uses sophisticated computer software to accomplish the same process, but much more quickly and at a lower cost. What is e-conversion? E-conversion, also called EDD, is the process of converting many different types of "native" electronic files into a single, common type that can be used for a wide variety or purposes. The most common technique converts electronic files (email, desktop publishing, spreadsheets, PDF images, etc.) to TIFF images for viewing and associated text files for searching. Many e-conversion processes also yield Metadata, which is file-level information about each file, such as its size, file extension type and last modified date. (See more on Metadata, below.) Intermediate How does Valora perform automated encoding? Valora's patent-pending software system makes several passes through a document set, identifying tell-tale document characteristics and extracting useful information. A trained Quality Control team reviews and enhances this information using custom-designed analysis tools that allow them to process documents at many times the speed of manual coding. The system then exports the data in the format requested by the client. How is Valora different from other auto-coding vendors? Most other vendors who provide "auto-coding" are providing a software-only service. That is, they run the images and text through their software and see what results. This "get-what-you-get" approach does not offer strong accuracy or complete coverage of all documents. This technique is sometimes called "dirty coding" since the raw software output is unenhanced by human review. All Valora's services pass through both computer processes and human processes to ensure the best quality results. In fact, we guarantee our population coverage at 100% and our accuracy at 98%. We are the only automated service provider to do so. How is Valora different from off-shore coding? Offshore coding is simply traditional, manual data entry that is happening somewhere outside the United States, typically India or the Philippines. The work is sent offshore to capitalize on cheaper labor rates than are available here in the United States, where we have minimum wage laws. The only difference between offshore coding and onshore coding is where the work is being performed, the familiarity of the staff with normal English business documents, and the resultant price of that work. There is no technological advantage in offshore coding. All of Valora's services are provisioned directly in our Burlington, MA (USA) headquarters. Nothing is ever outsourced and Valora does not use offshore labor of any kind. Additionally, both Valora's Burlington facility and most of our personnel hold "Public Trust" security clearance from the United States government. All of Valora's staff are American citizens. Advanced What is "strategic coding"? Strategic coding begins with a detailed analysis of your document population. This is usually done with a report that details what kind of documents you have and in what relative amounts. With this data, a strategic coding plan is produced that only captures the important fields per document type, rather than all fields all the time. This by-the-document approach ensures the best possible value for your limited coding budget. Valora's diagnostic tool is called the Document Distribution Report (DDR), which is created within 48 hours of receiving your data. With the DDR you can understand what's inside your population, how much of each type of document you have and ultimately, how best to process it. You said automated processing reduces costs. How can I use automation to the best cost advantage? By utilizing a DDR for paper populations and DDR + culling techniques (de-duping, near-duping, key word searching, etc.) on electronic populations, you can often get the processing costs under control. There is no need to fully code every document in a population. Instead, you can strategically code each document type to its best value. For example, you can code file folders for title only, but emails for title, date, author, recipient and copyees. Why would you code electronic files? What's wrong with metadata? Converted electronic files consist of three main pools of data: an image, text and metadata. On some files, metadata is accurate and useful information. On others, it is non-existent or wrong. Coding of electronic files ensures complete, accurate bibliographic information for each document, regardless of its source of origin. Below is a classic example of metadata versus coded data. Imagine you have a Word file. This could be a "native" hard drive file or an email attachment. It contains a letter from Susie Smith to Jimmy Jones. It is dated inside the letter as 3/3/2002, but the file was opened several times and resaved. It has the Re line text: "Pursuant to our meeting in Tampa." Compare the results below:
While the metadata is accurate for the Word file, it is not accurate for the document content and thus of very limited use. In contrast, TrueCodingTM, Valora's term for true, accurate coding of electronic files, captures the true author, title, etc. from the document content. With TrueCoding, you have 98% accuracy or higher in your coded output. |
|||||||||||||||||
 |
 |