Return arrow Back to Architecture & Engineering COE

Enabling a Smart User Experience Using Chatbots

Engineering

Digital transformation is fueling many kinds of evolutionary and revolutionary impacts to the digital Employee Experience within the Enterprise. Advanced technology capabilities in Artificial Intelligence, such as machine learning and natural language processing are increasingly being applied to the consumer space, and are now emerging as ripe for enabling smarter, more engaging experiences for enterprise employees.

Natural language processing (NLP) and chatbot capabilities are currently used in the consumer space for a variety of interactions ranging from answering customer questions to servicing requests for action and even ordering products.

Employee needs can be serviced in similar ways that offer an opportunity to re-imagine how employees engage with enterprise resources, and to leverage more modern interactions that are becoming increasingly familiar and expected. By taking a look at how NLP and chatbots function, and could be applied to enterprise contexts, we hope to inspire creative thinking on propelling digital employee experiences into new paradigms.

How does Natural Language Processing work?

Natural language processing allows spoken or written phrases to be analyzed by computers in order to determine the intent of the user. An intent can indicate the subject of the request or an action the user wants to take. The entities represent any pieces of information that provide proper context for a request or additional information required to complete an action.

In the example below, a manager has made a request for the amount of sick time used by a team member.

phrase_example

This phrase is run through a natural language processor to determine the intent and related entities. From this example, we can see that the intent has been identified in the context of the request along with two entities that help to focus the query to a specific user and with a specific conversion of time.

Training the natural language processing platform for accuracy

In order to determine intents and entities, the natural language processing platform must be provided a set of sample phrases that are mapped to a specific intent. This is a task that is performed by an administrator that is specifically trained to manage this process. As the administrator provides sample phrases for an intent they identify the entities that are present in the phrase.

Once an acceptable number of phrases have been provided and processed, the administrator initiates a training process within the natural language processing platform. This uses machine learning techniques to identify similar phrases for each intent.

As new phrases are provided by end users, the administrator can review them to ensure the natural processing platform is correctly identifying intents and entities. Corrective action is taken during this cycle to improve the accuracy of the platform.

Chatbots as virtual assistants

Allowing users to have natural conversations with a system opens up new possibilities for accessing content and services. Users can make direct inquiries in a natural way rather than rely on traditional navigational or search capabilities.

A chatbot is a technology that allows users to have natural conversations to access content and services. Chatbots typically take the form of a chat client, leveraging natural language processing to conduct a conversation with the user. Chatbots control conversation flow based on the context of the user’s requests and respond with natural language phrases to provide direct answers, request additional information or recommend actions that can be taken.

The diagram below provides a high level description of how a chat client could be used to leverage natural language processing to assist with access to content or perform data queries.

chatbot_workflow

Natural conversation adds to the user experience of accessing Enterprise services

The use of natural conversation will give users access to content and data in a manner that conforms to their needs. Users will no longer have to rely purely on search or navigation to find the content that is relevant to them. Enterprise services can be leveraged and made contextual for employees within the framework for a chatbot that maps those services to intents that NLP can extract.

An unprecedented range of machine learning and natural language processing capabilities exist today that can enable this type of smart experience within enterprise applications. This paradigm for fronting enterprise services with natural conversation has tremendous potential for revolutionizing the digital Employee Experience.