How SolrExperts ramped up the revenue 6 times by assisting ‘Expert Witness Profiler’ in their digital tool.
The Expert Witness Profile is the most comprehensive and cost-effective background report on an expert witness in the United States today. The product is used by lawyers, law librarians and paralegals practising in every area of law and by the widest array of organizations. From AMLAW 100 law firms to sole practitioners to national associations of attorneys to expert search firms and even by expert witnesses themselves.
Who is an Expert Witness?
An expert witness is a person whose level of specialized knowledge or skill in a particular field qualifies them to present their opinion about the facts of a case during legal proceedings.
The US Judicial Firm initially invested a lot of time in making the profile of an “EXPERT”. It took days in delivering The profile asked by the client. This process was exhaustive and required a high amount of precision work for very long hours. They wanted an indispensable tool to reduce the amount of manual work by decreasing the time taken to create a single profile and eliminate the tiring process of proofreading unimaginable pool of documents.
- It took almost 3-4 days creating a single profile
- A lot of documents were processed manually to find THE PROFILE details so they :
- Needed a searching tool to browse easily through the records
- Wanted NER to be implemented alongside to identify names in a document
- Creating a dashboard for their employees to generate THE PROFILE
- Analysing and examining a myriad of documents carefully to ensure no critical information is missed.
- Auto Crawling through private legal research databases, public, often obscure databases which most legal researchers would not think to query- or may not even be aware of.
- Extraction of the following data from the documents to create :
The Profile includes information about an expert’s:
Challenge (Daubert/Frye) history
Licensing and certifications (including verification)
Educational background (including verification)
Associations and memberships
Teaching and research
Patents, trademarks, and copyrights (if relevant) and
References in news, blogs, social networking, and more.
Named Entity Recognition (NER) labels sequences of words in a text which are the names of things, such as person and company names, or gene and protein names. We used Stanford’s NER, a part of CoreNLP a Java implementation of a Named Entity Recognizer. It comes with well-engineered feature extractors for Named Entity Recognition and many options for defining feature extractors.
The website offers a number of products ranging from a preliminary screening to an exhaustive report on an expert. The application developed with a responsive material design works flawlessly across all the platforms giving end users a premium experience.
The website hosts payment gateway for easier collection of fees. The application also caters to affiliates as a white label solution that gives a scope of complete GUI customization as per their needs
To simply state what SE did for the US law firm in the most uncomplicated way possible is that it enabled their employees to create a customisable page of the Expert Witness effortlessly similar to a Wikipedia page.
Distinguishing Features :
- A search bar where by typing the name of expert witness you obtain all the snippets of documents containing the name.
- Allowing the employees to select and reject the data from the snippets obtained and arranging them to create the profile – AUTOMATICALLY
- Auto extraction of data from private legal research databases, public, often obscure databases which most legal researchers would not think to query- or may not even be aware of
- Earlier they offered information which could be regarded as a week stale but now their reports are prepared with Real Time Data and are therefore always up-to-date.
- De-duplication of redundant information and disambiguating common names.
- Delivering the profile in 4 hours which earlier took almost 3-4 days.
- Reduced Time to Review a Case
Before – 3-4 days
After – 3-4 hours
- Improved Efficiency Of Work
Increased by 800%
All the data related to the cases were stored online. All the data was aggregated into a consolidated research platform, which allowed searching for Expert Witnesses, managing their data and doing research for the Profile. No need to dig into heavy paperwork to know the details of a case yourself.
6x Increase in revenue of the firm, due to increased efficiency and consolidated system.
NOT SO EASY
A lot of documents contain relevant information about an expert witness which are released every day by courts through their judgments. To state the numbers
- Total Cases: 1.3 million with a growth rate of 4.2%
- No of documents related to one case on average: 1.6 million
- Size of all these documents: 387 GB
Types Of Courts USA with the number of cases filed:
- The Supreme Court of the United States: 6,475
- The Federal Courts of Appeals: 60,357
- The Federal District Courts: 291,851
- The Bankruptcy Courts: 805,580
- The Federal Probation and Pretrial Services System: 137,410
Image Source: https://catalystsecure.com/blog/2015/08/how-many-documents-in-a-gigabyte-revisiting-an-e-discovery-mystery/