Electronic discovery (eDiscovery) is the process of collecting and analyzing electronic documents to determine their relevance to a legal matter. Office technology has advanced and eased the requirements necessary to create a document. As such, the volume of data has outgrown the manual processes previously used to make relevance judgments. Methods of text mining and information retrieval have been put to use in eDiscovery to help tame the volume of data; however, the results have been uneven. This chapter looks at the historical bias of the collection process. The authors examine how tools like classifiers, latent semantic analysis, and non-negative matrix factorization deal with nuances of the collection process.
TopA Brief History Of Legal Discovery
In countries that support discovery in the legal system, paper based discovery generally followed a simple pattern: identify the key people involved (initially referred to as witnesses and later as custodians), identify their support staff, get photocopies of the documents maintained by witnesses or from central filing systems, and send the boxes of documents to the legal team. The legal team would focus on the relevant custodians and thumb through the documents making judgment calls on the likely relevance of expense reports, budgets, and memorandums.
As volumes of documents grew, only larger law firms would continue making photocopies. In time, even those larger firms would be unable to keep up with the demands of photocopying. Without the means, ability, or desire to handle this manual job of photocopying, a market opportunity was created – the legal service provider. Specialists narrowed in on becoming litigation service providers. When a collection of paper-based documents was needed, the litigation service provider would step in and carry out the work that the law firm would not.
As self-motivated agents, the litigation service providers created a business driven by volume. Pricing for manual collections of paper documents were often priced by the page since the underlying cost was a combination of the staff required to handle the collection and the number of photocopies being made. This derivation of a service with revenues generated by volume set a pattern for future business models.
If a company found itself involved in a legal matter, it would often start by hiring a law firm. The law firm would then turn to their legal service providers to assist them in handling the matter. The litigation service provider welcomed advances in office technology that eased the creation of documents. With the gentle hum of the IBM Selectric on a secretary’s desk, the litigation service provider could count on more documents being printed and therefore more documents that needed to be photocopied.
Since the division of labor was cleanly split, law firms were not reviewing paper documents for relevance during collection. They would review the photocopies. Since the litigation service provider did not want to make legal judgments on document relevance, it behooved them to collect as much paper as they could and turn it over to the law firms for further evaluation. Thus a manual culling process carried out by legal experts was necessary. It was often too expensive to send people on repeated trips to make the physical copies of documents and they would collect and photocopy as much as possible. The drive to contain non-legal fees (like photocopying documents on repeated trips to the customer’s office) contributed to over-collection.