Answers to FAQs

Many technologies out there that claim to provide “AI” are, in fact, providing just rudimentary natural language processing (NLP). NLP is like sentence diagramming - a computer can identify the topics mentioned in an article, for example, or recognize that XYZ is a drug name. NLP is useful for keyword or key phrase searches. Natural Language Understanding, however, is more sophisticated. NLU derives additional information about an article or conversation, such as the sentiment of an author or speaker (e.g. supportive or contrary) or the subject of a research paper. NLU can categorize (or tag / label) conversations by abstract concepts such as “safety” or “efficacy”, even when those words do not appear within the actual source. This level of understanding only comes from sophisticated contextual knowledge built into the system through machine learning, algorithms, and ontology enrichments. It’s Sorcero’s “secret sauce”.

To effectively analyze large volumes of unstructured text at scale (such as millions of published articles), you need a text analytics engine that can parse long-form documents to identify topics, named entities, and metadata. Text analytics is the application of Natural Language Processing (NLP). Advanced text analytics engines also incorporate Natural Language Understanding capabilities and Machine Learning to identify themes, derive new data points (such as the sentiment expressed), or serve up recommendations and “next best” actions. Text Analytics is at the core of everything Sorcero does to help you research, explore data, collect evidence, and Know More, Faster.

We have large language models and a robust set of enrichments developed for the following therapeutic areas (TAs): Oncology, Cardiology, Hematology, Immunology, Infectious Diseases, Neurology, Rheumatology, Psychiatric, Immuno-oncology, and Gastroenterology. Don’t see your TA? Set up an inquiry and let’s talk. We are building new capabilities to speak your language all the time.

Yes! We never met a data source we couldn’t ingest, normalize, and unify into our scientific data repository. We provide all of our clients with the largest open access scientific data pool in the market, with 209 million publications, 213 million identity graph of authors and Key Opinion Leaders, 124,000 publishers (journals, congresses, preprints, and institutional repositories), 109,000 institutions, 1,001 biomedical ontologies, 79.6 million biomedical concepts, patents and grants information. Add to that any internal sources your team uses on a daily basis: CRM data, spreadsheets, powerpoint presentations, and more. Our Ingestum tool makes adding data easy. All of this data is enriched, meaning that we automatically tag topics and themes. There is no need to build your own data warehouse to get started with Sorcero. 

Insights represent the end state of observations that have been adjudicated and validated by medical affairs leadership and may come from either field observations, curation (e.g. developed from advisory board summaries, in-depth interviews, or medical meetings), or through internal strategy meetings where decisions are taken based on what is known by the leadership team. Insights are usually transformative in some way, driving action and finding their way into strategic imperatives. Ideally, an insight is more than one or two anecdotes; an insight should be new information, backed by evidence and validated by/in several sources. Sorcero surfaces intelligent groupings of information across data sources that MIGHT lead to an insight; Sorcero also helps medical affairs teams find, curate and share evidence, but only a medical affairs expert can elevate a collection of observations, evidence, and/or pre-insights to the level of “Insight”. Sorcero supercharges your ability to find emerging topics of conversation and other patterns, but a human in the loop is always necessary for that final recognition that “This is transformative.''