Knowledge Base

The Knowledge Base is a fundamental part of Analytica. It is the place where your information lives, already structured and ready to be used.
It should not be confused with a set of documents or files uploaded to the system (this type of data is intended as Files, and managed specifically in another section).
Our Knowledge Base already contains the information EXTRACTED from files (and from other data sources). In the IT sense, it is much more similar to an actual database.

Types

In the Types section, all the types already defined in the Knowledge Base are listed.

Creating a New Type

By creating a new type, you will be able to choose from a list of already defined templates.
Templates are maintained directly by the Analytica team, and represent common types from which you can take inspiration, and possibly modify them later.

When creating a type, its name is a fundamental point: it must appropriately describe the information you want to represent.

Examples of correct names to use:

  • Resume
  • Active Invoice
  • Opportunity
  • Balance Sheet
  • Bank Transfer
  • Real Estate Ad

Examples of incorrect names to use:

  • Document
  • Image
  • CV
  • Ad

You can notice that the incorrect names are very generic ("Ad" instead of "Real Estate Ad"), or extremely short (e.g. "CV" instead of "Resume").

Structure and Fields

Once a new type is created, it is necessary to model it according to the data you want to manage.
For example, if you want to collect information about books in a library, it is first necessary to think about what characterizes a book.
In this case, the type will be simply called "Book", and it should have some fields, such as:

  • title, which is the title of the book
  • author, i.e. the writer of the book
  • year, the year in which the book was published
  • isbn code, which uniquely identifies the book

which would already give me quite a lot of information about the book itself.
These 4 fields can immediately become 4 fields of the Knowledge Base.

Each field is represented by the following characteristics:

  • Field Name, which should therefore be, in our example, "title".
  • Type, which, for now, will be "Value". It represents a single value, as in our case.
  • Description, is the description of the field we want to model. So, in our example, it will be "the title of the book".

In this way, we are teaching Analytica what each field of our data is and what it should contain.
In the same way, we can also act on the "author", "year" and "code" fields.

More complicated examples will also be described below.

In the second part of field creation, it is possible to specify a Field Validation. If in the first part we described what the field represents, in this we will also define its type.
In the case of the book, the "title" and "author" will be validated as Text type fields, while the "year" can be validated either as Number, or as Date. In the case of Date, it is also necessary to specify the format in which the date is expected.
Furthermore, it is also possible to make a certain field mandatory, by checking the appropriate checkbox.
Analytica provides different types of validation, in order to cover almost all available cases.

Inserting Data into the Knowledge Base

Once your data type is defined, it is necessary to insert the data itself into the Knowledge Base.
To do this, it is possible to manually insert the data, through the "Knowledge Base" > "Data" section. In this section it is possible to see all the data already inserted, possibly modify or delete them.

However, Analytica connects the Knowledge Base with Files.
When a new file is uploaded in the Files section, in fact, it is first of all classified based on the list of defined Knowledge Base types. Secondly, the information defined in the various fields will be used to automatically extract the information.

So if I have defined a "Resume" type and I upload a pdf file that is a certain candidate's resume, this will first be identified as "Resume", and a new data will be created, in the Knowledge Base, with the "Resume" type, and the fields associated with it (which could be first and last name, email, work experience, etc.).

This processing comes into play every time a new File is created within the platform. So if, within the Chat, I upload a file, and I select the "Knowledge Base" mode, the file will be classified and, any of its information, will be made available in the Knowledge Base itself.

Furthermore, by exploiting the remote synchronization of folders, it is possible to automatically import even large amounts of information, already available in a folder on Google Drive.

Using Data in the Knowledge Base

All the information present in the Knowledge Base will be available in the various Analytica modules.
In the Chat, for example, I can request all the books by a certain author, or written before a certain year, and so on. A specific Agent will have the task of querying the Knowledge Base and making the information available.

Data Visualization

Analytica provides a dedicated view where it is possible to view all the data in the Knowledge Base.
Through this interface, it is possible to search for information by selecting the Type, delete and modify them.
To perform more in-depth or detailed searches, it is recommended to use the Chat directly, where it is possible to request information naturally.