Oracle Digital Assistant

We have several customers visiting our websites daily and they've had very common concerns. We used ODA version 21 to implement FAQs, and established a robust Question and Answer framework that utilized well history and data at large. Implementation on departments was notably smooth and the channeled conversations displayed an incredible messaging flow.

It's worked well on websites and apps for repetitive inquiries and frequently asked questions. It's suitable for personalized and more intense engagement with customers. I must admit that Oracle's machine learning ability and natural language processing is impressive which makes the AI conversations awesome. However, the AI is poor with "emotional IQ" but that's why you have the input from staff in the back-end.

Oracle Digital Assistant is very well suited to helping to solve basic, repeatable inquiries that require minimal technical expertise, helping to avoid committing human resources to very simple to solve problems, and keeping users happy without having to wait long times for a response.

For small organizations having a small number of business applications running for their organization, it would be as complex to integrate and study using ORM but for large organizations having numerous applications, working could go with this so that user single assistance for all applications.

Oracle Digital Assistant is very well suited for automating basic business processes and localized AI-enabled conversations in English.

It's less appropriate for manual tasks that require careful attention to detail. And it's also less appropriate for foreign languages, for instance, Spanish.

As mentioned in pros and cons, it depends on the use cases. Most of the normal chatbot use cases can be handled by ODA. If you want to build a chatbot with menu style or conversation style (it is not straight forward but it can be done), ODA can be perfect. If use cases are to also include AI with emotional intelligence to make conversational more interactive and also be able to detect through AI engine automatically what a person would like to do or perform or ask, then ODA may not fit for it. If you also are looking for virtual agents through tighter IVR integration then ODA may not be right. There are a few limitations around the number of words in the text to voice feature.

Well suited:

• Building a bot that needs continuous improvement.
• Conversion of FAQs into a suitable conversation bot.
• Building bots which need integration with third-party ERP applications or any other systems.
• Can integrate well with any third-party systems.
• Integration happens through API.

Less appropriate:

• I do not see an area where ODA can be inappropriate for building bots.