Need to identify stock mentions in Chinese social media or anomalies in oil rig logs ? LightTag's NER interface makes annotating spans easy, fast and accurate.
Whether you're preparing data for sentiment analysis or classifying radiological diagnoses, LightTag's document classification interface assists your domain experts in rapidly and efficiently applying classes.
If your specifying interactions between proteins and amino acids in a PubMed paper or relating quantities to prices in marketing material or anything else that needs to annotate relations, LightTag's Drag and Drop relation annotation interface makes it easy to do and manage.
Automate with AI
Bring your own models
You know your domain best, so upload your models, dictionaries or regular expressions and LightTag will present them to your annotators as suggestions. Increase annotator efficiency while validating your models.
LightTag learns from you
LightTag's MultiArmed Bandit based algorithms quickly learn high precision models that automate the easy things so your team can focus on where they bring the most value. Our deep learning based models adapt to your domain and language to generalize and provide broader suggestions to your team.
From Labels to Models
As your annotator interacts with the models LightTag learns you'll get reports on the model's accuracy. Once performance is excellent continue using the hosted model in production via a REST API.
Control, Measure and Improve
Get to Inter Annotator Agreement fast
Invite your team, tell LightTag how many labelers you'd like on every example and LightTag will dynamically assign work. Your labelers will never have to search for documents and you'll never have to manage work schedules. LightTag does it for you.
No code or devops needed
LightTag Just Works ™ out of the box. You get a managed solution, with daily backups and redundant servers to ensure technical issues never hold your team back. You can set up an annotation project in minutes without writing a line of code, configuring a database or worrying about network ports.
Know what's happening in your project
Get insights into everything that's happening in your project, the pace of annotation, individual annotator performance, agreement metrics and feedback on your models accuracy from your own team. All available in a convenient web based screen or through our API.