Author: admin

#ICYMI: http://bit.ly/2za837b

#ICYMI: http://bit.ly/2zbDU7b

 

Luckily, I was able to attend the #INBOUND17 Conference in Boston, powered by HubSpot in September and I came away with ONE BIG revelation.

“80% of businesses want chat bots by 2020”

Why wait? By then, you’re late.

I was struck by the predominance of the following message:  that customers now seek interaction that is so instantaneous that chats, powered by smart bots educated by big data and AI – are quickly becoming the new normal for online customer engagement.

Web browsing, form completion, email – all are too slow to produce answers in this attention-deficit-driven world. Closely related is the evidence that putting data to use intelligently via AI and machine learning produces an unquestionably better result than pure human might alone.

From sales inquiries, to support requests, to the creation of videos – AI is powering smart proactive engagement and content which is far more informed than that which can be created by humans. Imagine a product inquiry going to an inside sales human vs an onsite chat bot – which do you think can scan a knowledge base of millions of records of historical data faster to review all the possible responses to a question, to provide the best answer? The BOT.

An exaggeration? Perhaps. But, let me share some wake-up-calls from this message and tangible implications for my work at SAP in Marketing.

What are the common use cases/capabilities today?

AI/Chat bots:

* Provide smart personal assistants on sites
* Respond to Google search results – “I saw you were searching for …”
* Educate without annoyance of premature sales contact
* Create written content – “machine assisted content”
* Create video content – convert raw video into smart content
* Compose smart email replies

What best of breed bots and bot builders were shown at the event? 

Motion.ai (bought by Hubspot)

 

“Technology has changed the way we communicate. Today, people expect to interact with businesses when, where, and how they want.

Bots allow us to do exactly that, helping growing businesses scale one-to-one communications.

With HubSpot’s Motion.ai acquisition, soon anyone will be able to build a chatbot to communicate with customers on their terms — no technical skills required.”

Microsoft has QnAMaker

And also…

* Microsoft Xiaoice (China) 40 million users https://www.youtube.com/watch?v=dg-x1WuGhuI
* Atomic AI – helps marketers write better https://techcrunch.com/2016/12/12/atomic-ai-helps-marketers-write-better/
* Growthbot – chatbot for marketing and sales ps://growthbot.org
* Pathmatics – tells you what competitors do; monitor any other brand – every ad they run; how they ran them; https://www.pathmatics.com/
* Kemvi – bought by Hubspot; deep graph; social signals; writes email for the rep based on unstructured data https://www.hubspot.com/company-news/hubspot-acquires-kemvi-to-bring-machine-learning-to-sales-and-marketing
* Creating videos using AI :Magisto: machine assisted content. “Make inspiring video in minutes with the power of Magistos’s AI driven video editor. In 2017 your first hire should be a videographer and not a blogger.” https://www.magisto.com/

Winner for Thought-Provoking Presentation: Andy McAfee Spotlight: “Business Advice I Don’t Believe Anymore” – Andrew McAfee, Principal Research Scientist, MIT

Andy’s premise was that we are in the SECOND machine age, with all new rules, powered by data and intelligence. FIRST machine age conventional wisdom is all wrong. Some key points:

* Machines make better decisions than humans. Conventional wisdom is the HIPPO (highest paid person in the room) gets to make decisions, based on their years of experience and seniority. But with current big data analytics technology, “geeks armed with data” make far better decisions (MIT has documented). Disorienting for the HIPPOs, for sure, but critical that they learn to defer to this wisdom.

* Similarly, crowdsourced solutions deliver better results than those developed in internal, traditional processes. Crowdsourced platforms have created marketplaces to gather and reward such work. The wisdom of the crowd is real.

* Example: Quantopian consistently outperforms the market

* “Leveling Wall Street’s Playing Field, Quantopian inspires talented people everywhere to write investment algorithms. Select authors may license their algorithms to us and get paid based on performance.”

* Platforms spawn limitless innovations; in contrast, fixed products by their nature are more limited in their ability to incorporate innovations.

* Uber, ClassPass, AirBnB are platforms that constantly learn and consistently incorporate knowledge into their offering; you buy access to the platform and its ever-evolving services and not a “product.” In contrast, a product by nature is fixed and requires a different, slower process to incorporate intelligence.

Winner for Exciting Moment at Event: Michelle Obama Keynote!

Michelle was inspiring, talking about the importance of working through challenges and staying focused on what matters – satisfaction in your chosen path.

A few implications for my work at SAP in Marketing

The #INBOUND17 outside – in exposure helped me see the possibilities in AI/Chat – related work already taking place and made me more eager to test, learn and implement quickly. Examples:

* How fast are our B2B prospects moving beyond web browsing and form completion to expect chat and other personalized forms engagement on sap.com?
* What can we do to use AI to help create content (copy and videos) that leverages intelligence inherent in what has succeeded before?
* How can we use the mountain of stored intelligence in the SAP Community to help answer inquiries about support and training that come to sap.com?

Learn more

Here is a link to the #INBOUND17 website and collection of presentations and videos (scroll to bottom). Enjoy!! Lots of wisdom here and a great conference to attend.

  http://bit.ly/2zc2W6o #SAP #SAPCloud #AI

The following tutorial shows how to run a simple Spring Boot RESTful service on SAP Cloud Platform.

About Spring Boot

Spring Boot allows you to quickly setup and run production-ready Spring based Java applications. With only a little configuration you are able to run different scenarios such as RESTful services, JPA-backed persistence, etc. For more on Spring Boot continue here.

Sample application

Our sample application supports only one GET operation to /test-api/products

returning a JSON array of three different products.

You may retrieve the code from the GitHub repository https://github.com/pliegl/sample-rest-product-service

Application walkthrough

You only need four little artifacts, to get your application up an running. We use Maven to take care of the project dependencies.

pom.xml

The necessary dependencies are configured in pom.xml. We are using two different profiles – one to run the application standalone on the local machine and one for running it in a Tomcat 8 on SAP Cloud platform. dev true org.springframework.boot spring-boot-starter-web prod false org.springframework.boot spring-boot-starter-web org.springframework.boot spring-boot-starter-tomcat org.apache.tomcat.embed tomcat-embed-el logback-classic ch.qos.logback org.springframework.boot spring-boot-starter-logging org.springframework.boot spring-boot-starter-tomcat provided

SpringRestApplication.java

Our main class. The @SpringBootApplication and @ComponentScan annotations are sufficient to let Spring Boot know, what to do. * Sample RESTful application */ @SpringBootApplication @ComponentScan(“com.ecosio”) public class SpringRestApplication extends SpringBootServletInitializer { //Make sure that everything for .war deployment is there @Override protected SpringApplicationBuilder configure(SpringApplicationBuilder application) { return application.sources(SpringRestApplication.class); } public static void main(String[] args) throws Exception { SpringApplication.run(SpringRestApplication.class, args); } }

Product.java

Defines the structure of our products. We use Lombok to autogenerate getters and setter on a byte code level. /** * Sample product structure. Note that getters and setter are auto-generated by Lombok */ @Data @Builder public class Product { private String id; private String name; private BigDecimal price; }

ProductController.java

Defines our REST API. Currently, only a single GET method is supported. Additional operations such as POST, PUT, etc. may easily be added using the pertinent annotations and respective methods. /** * Serves a list of sample products */ @RestController @RequestMapping(“/test-api”) public class ProductController { private List products = new LinkedList(); @RequestMapping(method = RequestMethod.GET, value = “/products”) public List getProducts () { return products; } @PostConstruct private void loadDummyData() { products.add(Product.builder().id(“1”).name(“Apple”).price(new BigDecimal(“0.7”)).build()); products.add(Product.builder().id(“2”).name(“Pear”).price(new BigDecimal(“0.8”)).build()); products.add(Product.builder().id(“3”).name(“Orange”).price(new BigDecimal(“0.4”)).build()); } }

With these four little Java/XML artifacts we are ready to go.

Running the application

To run the application locally, run the following command on the command line: mvn spring-boot:run

To prepare the artifacts for deployment on SAP Cloud platform, run the following command on the command line mvn -P prod clean package

The final artifact sample-rest-product-service.war is located under the /target folder.

Deploying it on SAP Cloud Platform

* Login to SAP Cloud Platform on https://account.hanatrial.ondemand.com

* Chose JAVA Applications and then Deploy Application

* Select the previously generated sample-rest-product-service.war file and provide a name for your application. Select Java Web Tomcat 8 as runtime. Leave all other parameters unchanged.

* Check the status of your application
* In order to view the result of the GET request, click on the link as outlined below

The link shall look as follows: https://samplerestprodupXXXtria.hanatrial.ondemand.com/sample-rest-product-service/test-api/products

whereby XXX is specific to your SAP cloud platform account. Use a tool like Postman, to access the RESTful service. A regular Web browser will do as well for the simple GET request.

Summary

With only a few Java classes and clicks you are ready to deploy your first RESTful app into SAP Cloud Platform. http://bit.ly/2zbDEFf #SAP #SAPCloud #AI