Businesses around the world prefer to manage the content for their usage centrally. This content could be related to different functions/processes such as sales, marketing, policies, training etc. Once this content is available in a single repository, web sites are generally developed for the end users to access this content.
Key challenge is the ease of access to this content whenever user needs it, wherever user needs it. Apart from the websites, what if channels like intelligent chatbot is available for end users to search, fetch and access the content. Can this chatbot be seamlessly integrated with the commonly used tools like Microsoft teams or Slack.?
Oracle to the rescue
Oracle is the worldwide leader in providing SaaS and PaaS based solutions which spread across Customer Experience, ERP, HCM and more. Apart from these key solutions, Oracle also provides peripheral cloud services for Content Management and Conversational AI (chatbots) as well.
Oracle Content Management provides capability to create, publish and manage various types of content (documents, videos, images etc.).
Oracle Digital Assistant provides an AI platform to create conversational experiences for business applications through chat and voice interfaces.
Let’s have a look at how these two cloud services can work in tandem to provide seamless experience to access the content easily and quickly.
Oracle Content Management (OCM)
Oracle Content Management is a cloud-based content hub. It is a PaaS service provided by Oracle Cloud Infrastructure (OCI). It offers powerful collaboration capabilities to streamline the creation and delivery of content and improve employee engagement.
Organizations can create a repository of content which can be consumed by various users. This content can be published to different channels like websites, mobile, chat etc.
Key Solution Components are:
- Repository: Repository is a logical storage location for all the assets. It is an entity which manages all the files/folders in a structured order.
- Asset: An asset can be a content item that represents an individual piece of content, such as product literature, compliance documents, a blog post, case study, or a digital asset that represents an image, video, or other type of media.
- Taxonomy: A taxonomy is a hierarchy of categories to allow asset categorizations and help users find assets. It represents how content across organization is defined and classified. For example, define taxonomies for products, branches, compliance type, roles or any other hierarchy of subject categories that is relevant for your organization.
- Integration: Oracle Content Management provides REST APIs for content delivery. These APIs can be used by consuming systems, for example chatbots to search, fetch and display the content.
Oracle Digital Assistant (ODA)
Oracle Digital Assistant enables development of chatbots which understands natural language and can be interacted through voice or text. ODA powered by Natural Language Processing (NLP) can understand the user query and respond appropriately as per the skills it is trained for. ODA has API based integration capabilities which can fetch information from various sources.
- Natural Language Processing: Allows creation of skills to cater to the user’s inputs and commands. The Digital Assistant takes care of the processing using inbuilt algorithms to understand the inputs.
- Conversational flow and context: The context of the chat is maintained and based on the user input, the appropriate flow is invoked.
- Enterprise Integration: Custom components can be created to integrate with multiple applications using APIs to fetch data and send data to the users.
- Multi-Channel Support: Digital Assistant can be integrated with various channels, such as website, MS Teams which can carry the conversations back and forth from users on various messaging platforms to the digital assistant and its various skill bots.
Functional Flow
Oracle Digital Assistant integrated with Oracle Content Management can provide seamless access to content stored in Cloud.
The following example flow suggests searching and accessing various documents by different means.
- User can search documents by Categories.
- User can search documents by Name.
- User can search documents by Metadata (Content/Author)
Search by Categories
The user has an option to search by category hierarchy by choosing the category and then the Subcategory. System displays all documents under the Subcategory hierarchy as hyperlink which user can click and download.
Search by Document Name
The user has an option to search the document using document name. The system searches the documents in the entire repository based on the document name and shows the documents as hyperlinks which user can click and download.
Search by Document Metadata (Content/Author)
The user has an option to search the document using document metadata. The system searches the documents in the entire repository-based document content, document author and other metadata and shows the documents as hyperlinks which user can click and download.
Architecture
The following diagram depicts the typical ODA architecture.
In this solution, ODA integrates with OCM to read metadata of the documents based on user inputs.
A skill is created in ODA with different intents such as “List All Categories”, “Search Document By Content” and “Search Document by Name” for each of the options that user can choose. Based on the user selection the respective dialog flow is executed.
System components built in node.js has the logic to integrate with OCM to retrieve data based on user inputs which is used by ODA to display as results.
ODA has inbuilt capabilities for authentication using OAuth which is setup to enable authentication for APIs of external systems.
OCM has built in Content Delivery APIs which are used to search documents, retrieve metadata of document to generate clickable URLs for document download.
Below are the APIs which are used for this integration.
Get Taxonomy:
https://<<OCM_Host>>/content/published/api/v1.1/taxonomies?channelToken=<<channelToken>>&expand=children
Get Categories:
https://<<OCM_Host>>/content/published/api/v1.1/taxonomies/<<TaxonomyId>>/categories?channelToken=<<channelToken>>
Get Documents (By name, by metadata):
https://<<OCM_Host>>/content/published/api/v1.1/items?channelToken=<<channelToken>>
This skill can be used in different channels such as MS Teams, Intranet, Slack for use by end users. This skill can also be integrated with any other ODA based chatbot such as Oracle CX Sales in MS Teams which will provide a single window for end users for all conversations related to CRM.
Benefits
Integration of OCM with ODA provides the following benefits:
- Content Adoption: Access to content easily and quickly over chat channel to increase the usage of content.
- Increased Productivity: Employees can work more efficiently by accessing content at their fingertips.
- Single Source: One repository for content across different departments.