Chatbots and Conversation-Based search interfaces

A different navigational experience:  Instead of finding information via a search tab or drop-down menu, chatbots may open the door for conversation-based interfaces. And, companies can use the resulting feedback to optimize websites more quickly. The effect may be similar to the shift away from “like” buttons to more granular emoji-based reactions. We’re entering the age of context—a simple thumbs-up won’t cut it when we expect digital interaction to be more human and less text-based.

When thinking about the future of organic search,  chatbots could be poised to become a much greater part of the consumer search experience.  Bing has been testing chatbots directly in paid and organic search results, as shown below.

 

 

Similarly, Google has developed a new service called Chatbase, which provides analytics and suggestions for how to “fix” bot experiences to make them better for users. First of all, it is tapping into one of the biggest recent developments in the world of messaging apps: chatbots, or automatic response systems that sit on top of existing messaging platforms like Slack and Facebook’s Messenger. These tap into innovations in areas like natural language processing and machine learning to provide information and much more to users in a chat messaging format — a new way for users to interact with organizations and apps.

While chatbot integrations have been in the news over recent months, most people outside of the Seattle area won’t have seen them in action or truly considered how such integrations could be used.

For instance, if chatbot integrations within search results become a future reality, they could be used to carry out the following without ever leaving search results:

  • Book a test drive
  • Engage with customer service
  • Order products and services

The possibilities are vast and shine a light on the importance of APIs and data integrations to enable the next generation of consumer interaction.

If Bing’s testing is successful, and we see chatbots roll out in search results. Getting brands to a point where they can leverage the technology is going to be a challenge never before experienced by owned performance and marketing teams.

Conversation-based engagement creates a feeling of connectivity and can improve customer relationships with a brand. Marketing and sales departments can capitalize not only on the data collected via chatbots but from the ability to create highly effective campaigns that improve customer loyalty and edge out competition.

What brands need to do to prepare for such shift?

Send us an Email to for a strategy paper on this.

 

How the CognitiveEngage-ServiceNow integration revolutionize Customer Service

How the CognitiveEngage-ServiceNow integration can revolutionize Customer Service

Chatbots have arrived and how. We see them everywhere – for social media, websites and even in business-business conversations. It’s time they made an appearance and impact in internal communications as well!

Problems facing the modern HR

Every time you see an HR executive grimacing over the phone to some employee, you can guess at at-least some of what is ailing him/her. On any regular working day, the HR executives will be subjected to hundreds of calls coming his/her way concerning various problems an employee is facing. Navigating and resolving each of these calls is not only exhausting but can sometimes be unrealistic. Imagine having to scan through piles of catalogs to answer simple questions about an employee’s personal records or some IT problem. This invariably leads to exaggerated turnaround times per resolution, earning the ire of the user/employee to whom the question may seem to be a simple case of forwarding link over an email. Another problem with unstructured internal communication is that it is impossible to track these calls, such that the user loses track of the status of his/her query leading to a sense of confusion and frustration on both sides.

Enter ServiceNow

AI-powered ServiceNow is now integrated with the snowBOT designed to come up with an integrated solution for HR departments to capture requests through tickets, but not make it complicated for the end user.

With snowBOT, users are now able to:

  • Create incidents, service requests and search knowledge articles
  • Get quick reports and summaries (such as a status of tickets, pending approvals, etc)
  • Handle other miscellaneous requests

What’s more, today applications like CognitiveEngage can be used to create bots integrated with ServiceNow to further ease the process. Handling customer issues from the HR desk is quickly becoming a cakewalk!

The CognitiveEngage-ServiceNow Integration

Welcome to the world of seamless HR operations, courtesy a smart integration of the CognitiveEngage and ServiceNow applications. Here’s what you can do:

  • Handle General Q&A: Cognitive engage bot can respond to the frequently asked questions by the customer. These responses can be trained in IBM Watson Assistant services.  These questions are basically the requests which don’t require any incident to be created in Service Now. For e.g. A customer asking the steps to reboot a printer to the agent and agent can guide the user instantaneously.
  • Submit a New Incident: The Bot can create an Incident by capturing all the information from the customer. The basic information can be
    • Name
    • Address
    • Contact Information
    • Short Description
    • Detailed Description

The bot can understand the category of the request with the above information and submit the incident.  The customer will get the notification about the incident.

  • Report on Status of incident: A customer can ask for the status of the incidents at any time. Cognitive Engage can be integrated with Service Now API’s to fetch the status of the incidents. The bot can also show the list of incidents for a particular customer.
  • Enable Notifications: Customer can opt for push notification from the app through SMS or email.
  • Contextual Search for related incidents: Contextual search helps customers deflect or quickly resolve their issues without involving the service desk operators. Contextual search can also include results from other sources such as a service catalog, allowing a user to directly order a catalog item from a search.
  • Knowledge base search: The knowledge base search is basically searching for issues which are not trained as part of Watson training but searching in companies’ document repositories. This is a semantic search and not a simple keyword search.
  • Call back support for urgent incidents: In case the customer wants to talk to a customer care agent for any urgent incident, then the bot can notify any available customer care representative to make a call to the customer. Bots are constantly online and respond instantly to customers’ questions in real-time. If the Bot is unable to answer the question successfully, it can either route the conversation to an agent or if the entire team is unavailable, it has the ability to capture the customers’ details and pass the details over to the correct support agent. Acknowledge your customers instantly.

What’s in it for you: Benefits

  • Reduces time wastage: Gone are the days of long training sessions for all members of the HR every time a new technology comes into being. This new integrated Bot interface is a one-stop solution to handling almost all customer queries.
  • Powered by NLP: CognitiveEngage bot can understand natural language without having to rely on specific commands or structured text.
  • Instant customer acknowledgment: Receive instant replies instead of waiting for emails etc, or being asked to create tickets. Hence improved support quality that ensures customer satisfaction.
  • Increased HR productivity: Now, the HR can devote time on other urgent tasks, ones that need human intervention rather than focus on trivial questions, which results in increased ROI for the department.

As claimed, using chatbots like CognitiveEngage and ServiceNow can reduce call volumes to help desks by 15-20 percent, using the standard argument that frees human resources to handle more difficult inquiries. This not only boosts HR productivity and ensures profitability, but also improves corporate culture by ensuring seamless resolution of queries and lesser dissatisfaction level for agents and customers alike. That’s how CognitiveEngage-ServiceNow integration can revolutionize customer service.

App Connect Enterprise application as Microservice !

Background

The organization pursuing digital transformation must embrace new ways to use and deploy integration technologies, so they can move quickly in a manner appropriate to the goals of multicloud, decentralization and microservices. Challenges of a centralized ESB pattern are

  • Deployment changes could potentially destabilize other unrelated interfaces running on the centralized ESB.
  • Servers containing many integrations had to be kept running and patched live wherever possible.
  • Topologies for high availability and disaster recovery were complex and expensive.
  • For stability, servers typically ran many versions behind the current release of software reducing productivity.

Solution

A cloud-native application architecture lets developers use a platform as a means for abstracting away from underlying infrastructure dependencies. Instead of configuring, patching, and maintaining operating systems, teams focus on their software. By adopting cloud native architecture, the components such as deployment, delivery, resources, security, operations, routing are taken care by the cloud computing platforms. This architecture focuses on Dev ops (automation), Microservices, Containers.

Automation Flow

App connect enterprise is docker supported. The docker ACE runtime is lightweight. We can download the base ACE docker image from the docker hub. Base ACE docker image can also be built from the docker repository provided by ACE. Updated docker image can be built on top of the base ACE image with the updated code/configurations.

Following steps enable us to develop the services in ACE in microservices architecture with Dev ops automation.

Download the ACE docker repository https://github.com/ot4i/ace-docker. Keep it in an ubuntu system. It provides the docker files which are used to build the docker images. Build the base docker image from the provided repository. Jenkins server can be used to enable the automation. When Jenkins job is triggered, the source files are downloaded from the Git repository into the Jenkins repository. Build the bar file with the updated source files. Now, push the updated bar file into the sample folder of ACE repository available in the ubuntu system. Build the updated docker image on top of the base image. Deploy the docker image to the kubernates cluster set up in a IBM cloud or AWS or Google cloud.

Advantages

Docker container ensures effective isolation, resource sharing, Operational simplicity, Portability. A single service or a set of related services can be grouped together into a single docker container. Code changes, fixed configuration, environment configuration, runtime are updated in each container deployment. With the automation process, changes can be deployed to Test, Staging and production environments quickly and easily. Kubernates cluster makes the services highly scalable and highly available.

How Cognitive computing Is Revolutionizing Medical Education

 

A recent Johns Hopkins study suggests more than 250,000 people in the United States die every year from medical errors – other studies claim the numbers to be as high as 440,000.

Medical errors are the third-leading cause of death in the United States alone. Human factors including errors in diagnosis, gaps in history taking, and practitioners misinterpreting the initial information data encountered, play a major role.

GP or medical exams are designed to test the application of knowledge in the clinical context rather than just knowledge per se. It is designed to assess how a candidate integrates their applied knowledge and clinical reasoning, when presented with a range of clinical scenarios. It allows a candidate to demonstrate their clinical skills, communication skills and professional attitudes in the context of consultations, patient exams and peer discussions. It is a clinical consulting performance assessment.Let’s see how cognitive computing is revolutionizing medical education by the case study of a healthcare company.

Meksi, a specialized technology healthcare company, have identified that the current methods of assessing medical student competencies through case study role plays is entirely manual. It is time consuming, expensive and is not standardized.

That is why Meksi uses simulation techniques in a variety of learning, training and assessment scenarios. The adoption of simulations as a viable learning technique has sparked an evolution both in the teaching of medicine and how trainees and junior doctors develop their essential consultation skills.

With the ever-evolving nature of quality patient care, doctors not only have to master the knowledge and procedural skills, but also the ability to effectively engage with patients, relatives, and other health care providers, while coordinating a variety of patient care activities.

Meksi ensures that students and medical professionals, can prepare themselves efficiently against a pre-determined

benchmark in a simulated consultation environment before encountering them in real-life scenarios.

These simulated consultations provide trainees with the opportunity to learn and re-learn the processes as often as necessary. They are given the opportunity to correct mistakes and refine practices, enabling them to bridge the gap between their theoretical knowledge and practical real-world scenarios.

IBM Watson Artificial Intelligence

Meksi is revolutionizing medical education by maximizing learning outcomes through cognitive computing basically  using IBM Watson AI. Its platform allows medical professionals and students, to build competency in areas such as history taking, physical examination, diagnosis, ordering & interpreting investigations, as well as clinical management & communication with patients.

Leveraging the Machine Learning and NLP (Natural Language Processing) capabilities of IBM Watson, the Meksi platform creates a Virtual Patient and Case Study based clinical examination model that allows the assessment of the candidate’s medical knowledge, clinical skills and professional attitudes, for the safe and effective clinical practice of medicine.

The Virtual Patient

At most institutions, medical students learn communication skills through the use of standardised patients (SPs), but this is time and resource expensive. The use of Virtual patients (VPs) however, offer several advantages over SPs and can be used to teach medical students both history taking and communication skills.

This is important because understanding the patient’s actual input is an essential part of any diagnosis and allows for the disambiguating between different context, which itself, can be regarded as the real test.

For example, suppose a patient is suspected of having pneumonia. After three days on antibiotics, the patient hasn’t really improved. The Meksi system, having absorbed thousands of previous cases, enables the doctor with clinical skills, to identify whether the trajectory is normal or unusual, and whether a different medicine or course of action is required.

Asking these specific questions to understand the patient history and context at the right time is critical for successful diagnosis – the Meksi IBM Watson platform allows this to occur.

Meksi has created a virtual patient system leveraging AI driven conversations, to simulate interactive patient scenarios. This helps in reinforcing basic medical knowledge while solving a specific type of medical case with clinical reasoning.

Artificial Intelligence Based Assessment & Scoring Model

Progressive goal setting and feedback loops can help medical students to rapidly improve their skills and capabilities.

The Meksi platform provides this with its assessment model based on three different quotients axis – academic, behavioral, and test taking.

The assessment model evaluates…

  • Communication & Rapport – the ability to establish rapport and to communicate effectively with the patient in a pleasant, clear and logical manner, using appropriate communication skills and
  • History Taking – the ability to take a relevant and organised history, following appropriate cues and eliciting positive and negative details important to the assessment and management of the
  • Physical Examination – the ability to perform an appropriate and systematic examination that is appropriately focused and not overly inclusive. Candidates should be able to detect physical examination findings accurately and interpret them
  • Diagnosis – the ability to make an accurate diagnosis based on interpretation of the history, physical examination and
  • Management – the ability to manage the issues raised in the case. This may include immediate management (eg emergency measures), short-term management (eg safety-netting for the patient) and long-term management (eg prevention of recurrence), and preventive health.

The medical practitioner’s patient notes are assessed leveraging Natural Language Processing, that interprets techniques which include the extraction of facts, dosage information, plus the complexity of decisions made by the medical practitioner.

The Meksi AI enabled platform, has been developed by a team of the world’s most respected medical, digital and creative minds.

Company: Meksi Website: www.meksi.com

Industry: Healthcare

Solution: IBM Watson AI

How the CognitiveEngage-ServiceNow integration can revolutionize Customer Service

 

Chatbots have arrived and how. We see them everywhere – for social media, websites and even in business-business conversations. It’s time they made an appearance and impact in internal communications as well! So, it’s time to take your customer service a level-up using chatbots like CognitiveEngage and ServiceNow.

Problems facing the modern HR

Every time you see an HR executive grimacing over the phone to some employee, you can guess at at-least some of what is ailing him/her. On any regular working day, the HR executives will be subject to hundreds of calls coming his/her way concerning various problems an employee is facing. Navigating and resolving each of these calls is not only exhausting but can sometimes be unrealistic. Imagine having to scan through piles of catalogs to answer simple questions about an employee’s personal records or some IT problem.

This invariably leads to exaggerated turnaround times per resolution, earning the ire of the user/employee to whom the question may seem to be a simple case of forwarding link over an email. Another problem with unstructured internal communication is that it is impossible to track these calls, such that the user loses track of the status of his/her query leading to a sense of confusion and frustration on both sides.

Enter ServiceNow

AI-powered ServiceNow is now integrated with the Cognitive Engage bot designed to come up with an integrated solution for HR departments to capture requests through tickets, but not make it complicated for the end user.

With Cognitive Engage bot, users are now able to:

  • Create incidents, service requests and search knowledge articles
  • Get quick reports and summaries (such as a status of tickets, pending approvals, etc)
  • Handle other miscellaneous requests

What’s more, today applications like CognitiveEngage can be used to create bots integrated with ServiceNow to further ease the process. Handling customer issues from the HR desk is quickly becoming a cakewalk!

The CognitiveEngage-ServiceNow Integration

Welcome to the world of seamless HR operations, courtesy a smart integration of the CognitiveEngage and ServiceNow applications. Here’s what you can do:

  • Handle General Q&A: Cognitive engage bot can respond to the frequently asked questions by the customer. These responses can be trained in IBM Watson Assistant services.  These questions are basically the requests which don’t require any incident to be created in Service Now. For e.g. A customer asking the steps to reboot a printer to the agent and agent can guide the user instantaneously.
  • Submit a New Incident: The Bot can create an Incident by capturing all the information from the customer. The basic information can be
    • Name
    • Address
    • Contact Information
    • Short Description
    • Detailed Description

The bot can understand the category of the request with the above information and submit the incident.  The customer will get the notification about the incident.

  • Report on Status of incident: A customer can ask for the status of the incidents at any time. Cognitive Engage can be integrated with Service Now API’s to fetch the status of the incidents. The bot can also show the list of incidents for a particular customer.
  • Enable Notifications: Customer can opt for push notification from the app through SMS or email.
  • Contextual Search for related incidents: Contextual search helps customers deflect or quickly resolve their issues without involving the service desk operators. Contextual search can also include results from other sources such as a service catalog, allowing a user to directly order a catalog item from a search.
  • Knowledge base search: The knowledge base search is basically searching for issues which are not trained as part of Watson training but searching in companies’ document repositories. This is a semantic search and not a simple keyword search.
  • Call back support for urgent incidents: In case the customer wants to talk to a customer care agent for any urgent incident, then the bot can notify any available customer care representative to make a call to the customer. Bots are constantly online and respond instantly to customers’ questions in real-time. If the Bot is unable to answer the question successfully, it can either route the conversation to an agent or if the entire team is unavailable, it has the ability to capture the customers’ details and pass the details over to the correct support agent. Acknowledge your customers instantly.

What’s in it for you: Benefits

  • Reduces time wastage: Gone are the days of long training sessions for all members of the HR every time a new technology comes into being. This new integrated Bot interface is a one-stop solution to handling almost all customer queries.
  • Powered by NLP; CognitiveEngage bot can understand natural language without having to rely on specific commands or structured text.
  • Instant customer acknowledgment: Receive instant replies instead of waiting for emails etc, or being asked to create tickets. Hence improved support quality that ensures customer satisfaction.
  • Increased HR productivity: Now, the HR can devote time on other urgent tasks, ones that need human intervention rather than focus on trivial questions, which results in increased ROI for the department.

Using chatbots like CognitiveEngage and ServiceNow can reduce call volumes to help desks by 15-20 percent, using the standard argument that frees human resources to handle more difficult inquiries.

This not only boosts HR productivity and ensures profitability, but also improves corporate culture by ensuring seamless resolution of queries and lesser dissatisfaction level for agents and customers alike.

Discover valuable insights from Call centre data

The next big thing in data analytics is Speech recognition and secret powers are turning spoken information into actionable intelligence. There is currently a plethora of recordings containing valuable customer and business insights that are simply falling off the table. Just recording isn’t enough anymore if you can’t access and analyze your recordings at scale.

Machine learning based automation technology allows scalability through Automatic Speech Recognition by converting automated speech-to-text transcription mostly it can be used to discover valuable insights from call centre data.

Getting the transcription down quickly and accurately is just one piece of the puzzle to getting the most out of your customer conversations. Now you have to make sense of the data. This means pulling target keywords and phrases that provide you with information that can help you make business decisions and create a better customer experience. Manually pouring through the transcripts and making predictions is a decidedly cumbersome job that perhaps doesn’t produce the best results.

  • sentiment analysis to understand agent performance and customer experience
  • target keywords and phrases that provide you with information that can help you make business decisions and create a better customer experience.
  • predictive analytics to gather information about hot leads, Up-sell probability, potential churn, caller intent.
  • monitoring and automation for regulatory and compliance