Oracle Cloud Infrastructure (OCI) Data Labeling is a service for building labeled datasets to more accurately train AI and machine learning models. With OCI Data Labeling, developers and data scientists assemble data, create and browse datasets, and apply labels to data records through user interfaces and public APIs. The labeled datasets can be exported for model development across Oracle’s AI and data science services for a seamless model-building experience.
Upload documents, including PDF and TIFF formats, and add labels. These labels are helpful for scenarios like training custom document-classification models for classifying support tickets and automatically processing refunds based on customer explanations.
Developers can upload raw images, add labels, and highlight areas of images. By adding these labels to images, the resulting datasets can be used to train custom image classification and object-detection models.
Upload text-classification labels and use OCI Data Labeling to automatically identify key information in text. This labeled text can be used to train custom natural-language processing models for information extraction, intent classification, sentiment analysis, and more.
OCI Data Labeling provides custom templates and multiple annotation formats. Label data according to the needs of machine learning models. Annotate images, text, or documents in just three steps: Create a dataset by loading data, annotating it, and exporting it.
Export a snapshot of the annotated data record in JSON format to object storage. Access exported labeled datasets across Oracle’s AI and data science services, and integrate them into custom model-building processes without any transformations.
Use OCI Data Labeling on its own, or access it within other services such as OCI Vision and OCI Language. Developers and data engineers can assemble and label datasets and then easily reference them via OCI AI Services as part of a custom model-training workflow. Data scientists who prefer to build and train their own deep learning or natural language processing models can consume the labeled dataset through OCI Data Science.
Label datasets consisting of logos, popular clothing silhouettes and colors, types of products, and medical images. Use these labeled datasets for inventory planning, product categorization, shelf management, and medical diagnoses.
Label irregular images to create models that automate elements of product quality checks, defect detection, safety surveillance, and inventory management.
Label documents to make extracting valuable information easier for patient claims processing, medical report diagnostics, and cellular research.
Classify datasets including receipts, invoices, and POs for customer support chatbots and automated expense filing.
Tag groupings of words and assign labels. Labeled text datasets can be used for customer survey analysis, topic modeling, and customer support.
Praveen Patil, Principal Data Scientist
OCI Data Labeling service lets you assemble data into sets, create and browse datasets, and view records (such as images and text). The service has a rich user interface for applying, editing, and storing labels, and then collating the labeled data so that you can train and build custom models through Oracle AI and data science services. You can access these capabilities through the console and via APIs that let you build your own data labeling workflows.Read the complete post