HaystackID^® Discovery Intelligence:
Powering Next-Generation Discovery

According to the Electronic Discovery Reference Model (EDRM) ⁽¹⁾, artificial intelligence (AI) refers to the capability of machines to mimic aspects of human intelligence such as problem-solving, reasoning, discovering meaning, generalizing, predicting, or learning from experience. AI includes both unsupervised and supervised machine learning and has many other processes, such as natural language processing (NLP). In the context of HaystackID Discovery Intelligence, AI describes the automated discovery processes used to classify, categorize, summarize, predict, and provide insight and intelligence regarding ESI. This application of AI by HaystackID is accomplished through the expert application of proprietary statistical, rule-based, and algorithmic means. The potential of AI is realized in HaystackID Discovery Intelligence in areas ranging from detecting sensitive data (e.g., from PII and PHI to SSNs and License Plate Numbers) and identifying entities to support cyber breach responses as part of ReviewRight^® Protect™ services to the evaluation and classification of legal documents reviewers as part of HaystackID ReviewRight Match^® services. Three examples of AI-enabled services available as part of HaystackID Discovery Intelligence include:

Data science can be defined as the process of using algorithms, methods, and systems to extract knowledge and insights from structured and unstructured data. ⁽²⁾ As part of HaystackID Discovery Intelligence, HaystackID’s team of data science authorities have the expertise and experience to support the challenging and complex discovery tasks associated with the complete data science lifecycle. This lifecycle includes capturing, maintaining, processing, analyzing, and communicating intelligence to reach decision points and inform decision-makers.

According to IBM ⁽⁴⁾, machine learning is a branch of AI and computer science that focuses on using data and algorithms to imitate the way humans learn, gradually improving its accuracy. In HaystackID Discovery Intelligence, HaystackID leverages predictive coding, a term used to describe a technology-assisted review (TAR) process which involves using a machine learning algorithm to distinguish relevant and non-relevant documents based on a subject matter expert’s coding a training set of documents. ⁽⁵⁾

HaystackID Discovery Intelligence predictive coding is based on active learning. Active learning is a process, typically iterative, whereby an algorithm is used to select documents that should be reviewed for training based on a strategy to help the classification algorithm learn efficiently. ⁽⁶⁾ HaystackID Discovery Intelligence typically leverages a Continuous Active Learning^® (CAL^®) predictive coding protocol. In CAL, the TAR method developed, used, and advocated by Maura R. Grossman and Gordon V. Cormack, after the initial training set, the learner repeatedly selects the next-most-likely-to-be-relevant documents (that have not yet been considered) for review, coding, and training, and continues to do so until it can no longer find any more relevant documents. There is generally no second review because, by the time the learner stops learning, all documents deemed relevant by the learner have already been identified and manually reviewed. ⁽⁷⁾ Two examples of machine learning-enabled successes delivered by HaystackID as part of HaystackID Discovery Intelligence include:

While innovative technologies and techniques delivered by AI, data science, and machine learning are fundamental to discovery success, at the end of every process and practice, there is a requirement for human intervention or evaluation that can span the spectrum from document and source code review to deliverable authentication, validation, and verification. As the industry’s most experienced and proficient provider of secure document review services with more than 2,000 virtual projects completed and more than 20,000 trained reviewers supporting onsite and remote reviews, HaystackID’s ReviewRight offerings provide the expert and experienced human touch powering HaystackID Discovery Intelligence. HaystackID Discovery Intelligence managed review practicality in action includes:

HaystackID Discovery Intelligence Synergy in Action

One example of the synergy provided by the combined application of HaystackID Discovery Intelligence artificial intelligence, data science, machine learning, and expert reviewers is manifested in HaystackID’s Protect Analytics^™ offering and the promise, precision, power, and practicality it provides.

Protect Analytics is an exclusive set of technologies and processes that allows client data set analysis for sensitive information ranging from PII and PHI to data breach code anomalies. Enabled by a collection of proprietary workflows and proven tools, Protect Analytics can proactively or reactively help determine sensitive data concentrations, locations, and relationships to inform notification lists, exposure assessments, and discovery targeting.