Linguistic Pattern Matching

Linguistic Pattern Matching ("LPM") is a sophisticated new search technology that gives legal teams the ability to work much more quickly and thoroughly with large databases of scanned discovery documents and electronic files. In fact, when combined with computer-generated auto-coding, LPM represents the fastest, most precise way to analyze large document populations.

The technology which drives LPM identifies the linguistic patterns of words found on individual pages of text and uses those patterns to efficiently sort and link related documents -- just as a human would do, only thousands of times faster. It can improve a variety of everyday tasks and, since LPM databases are prepared electronically, they can be available for use virtually from the very start of the litigation, unlike manually coded databases that can take months to prepare.

Whether operating as a standalone search tool or in conjunction with common document management software, LPM is typically easy to learn. Invoking a search simply requires selection of blocks of text or even an entire page of text to use as the basis for the search. A search can also be based upon text found outside of the collection. For example, interesting passages of text residing in litigation support software, transcript review software or in briefs being examined in Word can be highlighted and used for comparison against the database. Or, in some cases, lawyers need only type a textual narrative which describes a document they're looking for.

Again, LPM analyzes the linguistic patterns contained in the comparison text and finds those pages in the database that contain textual patterns most nearly matching that text. Users do not need to use any complicated search syntax such as Boolean operators (e.g., "Jones AND Smith AND counsel"), proximity operators ("legal pre/2 advice"), or wildcards ("wom?n" or "spoliat*") - just the source or comparison text.

Unlike traditional Boolean searches where the most relevant documents can be scattered at random throughout the results, LPM results are ranked according to how closely the linguistic patterns in the documents match the source or comparison text, helping searchers immediately focus their attention where it is most likely to yield the best results.

Here are some ways that LPM can help legal teams perform critical tasks in litigation, beginning from the very start of discovery:

Consistent Assertion of Privilege

LPM provides a thorough way to check scanned documents and e-discovery to be certain that redaction has been done consistently. When claiming privilege on select documents, the party asserting the privilege certainly wants to avoid having duplicate copies being inadvertently produced.

With automated database creation, producing parties can use LPM to search a collection for privileged passages before the production is complete. Searches based upon passages from privilege logs will quickly identify potentially privileged documents that can be reviewed and appropriately flagged.

Easy search formulation and rapid results evaluation make this a far superior approach than Boolean searching, which can require substantial expertise and trial-and-error to refine searches. With redactions there are simply too many things to check to be able to frame multiple Boolean searches.

Electronic Spoliation Audits

The proliferation of e-mail and other digitized data takes the risk of inadvertent spoliation to a whole new level. Counsel can audit productions by opposing parties for instances of electronic spoliation, with the goal of obtaining sanctions, such as favorable jury instructions. When producing, counsel can audit their production with the goal of the audit being to make sure that there isn't any evidence in its own paper and electronic production that could indicate failure to entirely produce all of its electronic evidence. LPM provides a convenient method to do this because both scanned paper documents and electronic files reside in one database.

To start, users should examine paper records for which electronic files must at some time have existed in electronic form and then, using LPM, determine if those electronic records were in fact produced. The typical example is e-mail. If paper copies of e-mails exist, then there should also be electronic copies. Other examples of paper documents that almost certainly existed at one time in electronic form are PowerPoint presentations, Excel spreadsheets and Word documents, such as contracts, letters, memos, training guides, etc. LPM enables the legal team to zero in on those that may be relevant.

Locating Lost Documents

In the course of litigation an attorney or witness may simply "know" that they've seen a document and even have a dim recollection of its wording. With LPM, all the legal team has to do is type out the document as it's remembered and then use that text as the source for a search. Even if the source text is not exactly what was in the original document, it may still provide enough information to locate the missing document, without the need to resort to complex and time-consuming queries.

Finding E-mail Chains or Clusters

Electronic dissemination and replication of documents can create an information overload for the legal team. For example, each recipient of an e-mail can reply, reply to all, forward, blind copy, print, or copy and paste the e-mail into word processing or other applications, or even post it to a list serve with thousands of subscribers. One original can quickly multiply into chains and clusters of duplicates.

LPM lets lawyers and paralegals rapidly identify e-mail threads or chains resulting from a series of replying to, forwarding, or BCC'ing original emails. It also identifies situations where the contents of e-mails have been copied and pasted into other e-mails or other documents in the collection. Identifying duplicate and near dupes like this greatly simplifies the job of analyzing records so that precious time and money aren't wasted reviewing the same documents over and over again.

Provided by Joseph Howie.