We use a simple three level structure to organize data. Each customer is assigned to a single Organization and can manage his Datasets and Documents as needed.
Data Model Overview
The Organization is the top level of the data model. Each customer is assigned to a single Organization and thus isolated from the others. It is not possible in any way to access an organization other than the one a customer is assigned to.
An organization contains all data and parameters relative to a single customer. From a customer point of view, the Organization is transparent as there is no way to interact with this entity.
Each organization can contain several datasets (the number of datasets is limited depending on the customer plan). A dataset defines several parameters allowing Dictanova Engine to process data in a proper way.
- Unique name
- Default Language to be used for text and sentiment analysis
- Document Type and Business Sector to give a hint to Dictanova Engine on how to process documents
Documents can be imported in a dataset along with metadata fields. To get the most value from these metadata, each of them must be defined on the dataset level :
- Unique code
- Type (satisfaction score, number, date, ...)
- Constraints (bounds, allowed values, ...)
All documents imported in this dataset must be consistent with the defined model.
Refer to Dataset Data Model to get full details about the Dataset entity.
Datasets are containers that allow data importing. Each element imported in a dataset to be analyzed and stored is a Document.
A document is made up of three types of information:
- Text content to be analyzed
- Metadata consistent with the model defined on the dataset
- Parameters such as language and document type to give a hint to Dictanova Engine on how to process content
Refer to Document model to get full details about the Document entity.
It is not possible to create a single document in a dataset. It is mandatory to use the Import feature to send your documents for analysis.