Data Discovery is not what it once was.

The Nebula product team, consisting of industry experts, a team of data scientists, and software developers, have been at the forefront of providing clients the best solutions and tools in the industry to reduce costs and focus on relevant data faster, for decades – that’s right, decades. Nebula offers a full range of Machine Learning, Natural Language Processing, and Workflow Automation tools to help accelerate the analysis and review of information to help you get the most out of your data.

Paper documents tranforming to data.

Machine Learning

Leverage human expertise to automatically classify millions of documents in a matter of hours.

Nebula’s supervised machine learning, known as “predictive coding” in the legal industry, supports multiple workflows and methodologies and helps prioritize essential documents for review. As a result, Nebula can automate the classification of electronic data, drastically reducing the time required for legal review and saving customers millions annually. Our predictive coding technology gets smarter with every document reviewed and supports proven statistical methodologies to ensure a defensible workflow every single time.

Continuous Active Learning (CAL) is the go-to method of augmenting human document review to improve efficiency and accuracy, and control costs. Its simplicity and shallow learning curve help eliminate the barrier to entry often encountered with training by subject matter experts, who are often highly paid hourly resources. When applied to document review, Continuous Active Learning is very popular in scenarios where there is a desire or requirement to put human eyes on all or most documents at issue in a case.

Nebula includes proprietary CAL technology that allows for minimal human intervention and will continue to classify documents based on training from a limited number of participants.  Nebula’s powerful automation ensures experts are always seeing the next-most-likely-to-be-relevant documents for review, coding, and training. This continues until no more relevant documents are available for analysis.

Natural Language Processing (NLP)

Powerful language-based AI enables users to gain insights into data sets in ways never before available.

For example, visually locate and search for documents about named entities, such as locations, events, and key people, with the added ability to filter for documents containing critical language by analyzing the author's sentiment.

Nebula’s entity extraction engine is trained to recognize eight distinct categories of real-world entities then visually cluster documents referencing the same entities. This approach, based on semantic understanding rather than simple word frequency provides enhanced insight into the data, allowing users to isolate and retrieve relevant information or filter non-relevant material quickly.  Nebula can also uncover cover topics that might otherwise go unnoticed, giving legal teams an edge over the opposition.

Nebula’s sentiment analysis tools analyze tone at both the document and sentence levels. At its core, sentiment analysis applies Natural Language Processing techniques and computational linguistics to derive emotional attributes from text content.  By leveraging sentiment analysis, you can better understand how communications are perceived and help discern the author's tone and intent. This gives Nebula users an edge in contexts where more than just the words themselves matter, as in, for example, matters related to workplace harassment. Companies can use this to learn the tone of their employees to help determine if communications are positive, negative, or neutral, and help understand the behaviors and communication styles of employees and customers to identify trends and defend and identify bad actors.

Workflow Automation

Nebula’s workflow engine automates the routing and distribution of documents to streamline document review and maximize accuracy and defensibility.

Workflow eliminates the need to maintain static batch sets and manually transition records to different review teams. Instead, documents flow automatically through completely customizable paths based on an endless number of potential criteria: foreign-language documents to native speakers, privileged documents to senior attorneys, random samples of reviewed documents to quality control teams, and so on. Since documents cannot move on without meeting defined criteria, an additional layer of quality control is built in. Workflow works hand-in-hand with predictive coding to provide the most efficient review possible.