Artificial Intelligence and Machine Learning
Artificial Intelligence and its sub-component, machine learning, are technologies that enable computers to augment or replace human efforts. Within the legal industry, the concept of predictive coding is well-known and largely proven. Predictive coding is typically used to replace or supplement document review processes for the purpose of identifying responsive, privileged, confidential, or other document categories. We believe that the core competencies related to predictive coding are: a) detailed understanding of the algorithms used in this process, b) statistical expertise in quantifying the accuracy of the process, and c) testimony skills explaining and validating all of the related steps in using predictive coding.
Outside of predictive coding, there are also a great number of legal processes that can be supplemented by the accurate use of artificial intelligence. Basically, any process that can identify an appropriate training set that the artificial intelligence model can learn from can be automated. Tasks such as contract standardization, database investigation to identify fraud or other bad acts, asset identification and tracking, social media investigation, etc., are all candidates for advanced analytics and artificial intelligence.
Relational Database Analytics
Many components of technology used within the disputes and investigation space, such as electronic discovery, are well-known due to their frequency of use. Other components are more unique and are used with much less frequency. Our goal at Vista Analytics is to continue to develop these unique and custom technological applications and processes to: a) optimize traditional data queries and reports, and b) maximize artificial intelligence techniques. The investigation of large, structured databases is by its very nature a unique and custom process, and projects of this type have historically been lengthy and expensive. We believe we can successfully combine our technological offerings to achieve shorter project timelines, lower costs, and greater insights into data that could be missed when limited to human imagination and experience.
Traditional third party tools, particularly in the realm of disputes and litigation, are usually designed to automate solutions to common problems and identify particular data types. While Vista Analytics strongly believes in the continued use and optimization of these products, there are always new problem sets, unique data types, and innovative analyses that require custom programming and specific technology. Vista Analytics has proven experience in a broad range of relevant systems, but more importantly has the design expertise and requirements analysis that custom programming requires. Whether the need is specific functionality, high volume capabilities, throughput optimization, specific security requirements, or pieces of all these subsets, Vista Analytics can provide valuable expertise and vision.
High Performance Systems
The volume of data in the world is increasing exponentially. The technology to investigate and manage that data is also increasing at a rapid pace. While the ability to use millions of database records in litigation just two decades ago was daunting, developed techniques such as statistical sampling have allowed for the evaluation of large amounts of data. Today, it is possible to cost-effectively and rapidly investigate database sets of tens of billions of records.
Technology such as Hadoop clusters and in-memory databases can greatly assist in these investigation endeavors. Hadoop enables clusters of commodity workstations to work together and greatly increase computing power. In-memory eliminates any of the I/O related to a computer process, which is the slowest part of any computer operation. Maximizing these technologies can enable work to be completed in days rather than months, while still maintaining the redundancy and security of these systems.
Electronic discovery has become a commodity-based industry. There are simply a myriad of options worldwide to have information processed and hosted both professionally and cost-effectively. Vista Analytics realizes, however, that key questions within the industry remain. Such questions revolve around quality testing, exception handling and documentation, and throughput and validation of results. Vista differentiates itself by focusing on such key questions and offering options spanning ownership of the total process to segmented solutions around testing, documentation, and attestation.