Introduction

Since the internet boom a few years ago companies started to collect and save data in an almost aggressive way. But the huge amounts of data are actually useless if they are not used to gain new information with a higher value. I was always impressed by the way how easy statistical algorithms can be used to answer extremely complex questions if you add the component “Machine Learning”. Our goal was to create a web service that does exactly that thing: We realized Weather Forecasts using Machine Learning algorithms.

The Application can be split in four parts:

• The website is the final user interface to start a query and to see the resulting prediction.
• The server connects all parts and hosts the website.
• The database stores all important data about the weather.
• IBM Watson is used to calculate the forecasts with the data of the database.

In the following I will explain the structure more detailed and show how we developed the application.
Continue reading →

Welcome to our blog post ‘Moodkoala – An intelligent Social Media application’. The following provides an overview of our contents.

Contents

Introduction
       – The idea behind Moodkoala
       – Technologies overview
Technologies
       – Frontend and Backend
       – Bluemix Services
       – Liberty for Java
       – Natural Language Processing
       – Tone Analyzer
       – Language Translator
       – Cloudant
       – Mood analysis with IBM Watson
       – Mood Analysis with IBM Watson Tone Analyser
       – The Mood Analysis Algorithm
       – Embedding the Tone Analyzer into the Java EE application
       – Filtering Hate comments using Natural Language Understanding
       – Natural Language Understanding
       – Summing up text analysis
       – Google and Facebook API
       – Mood imaging analysis
Implementation
       – Implementing the mood analysis algorithm
       – Deserializing JSON Strings
       – Implementing the Natural Language Processing API
       – Implementing hate comment filter into the Java EE application
       – Set up Google Sign-in
       – Set up Facebook login
       – Mood imaging implementation
       – Configuration
       – Docker and IBM Bluemix
       – Gitlab CI
       – Jenkins
Discussion and conclusion
       – Discussion Moodkoala and 12 factor app
       – Comparison to other cloud providers
       – Conclusion
Continue reading →

Introduction

IBMs artificial intelligence ‘Watson’ on the IBM Bluemix platform offers a wide range of cognitive services like image and audio analysis among other things. During our semester project in the lecture ‘Software Development for Cloud Computing’ we integrated useful Watson services into a Telegram chat bot to provide a convenient form of direct access to those services right out of an ongoing chat conversation.

Our goal was to create an intelligent Telegram bot with services like Tone Analyzer for audio messages, Visual Recognition for analyzing sent images, a Speech-To-Text functionality to be able to get spoken text inside audio messages and a translation service. Working both with a Telegram bot and IBM Watson services we came up with the name combination ‘Botson’.

In this blog post you’ll learn the basics on how to create and configure a Telegram bot as well as why we recommend creating several bot instances when working in a team. Afterwards we’ll provide an overview of the IBM Watson services that we used, what you should keep in mind when creating them in IBM Bluemix and the problems that we were facing along the way. Another part of the IBM Bluemix chapter covers how we configured the Continuous Delivery service, why we struggled to connect our existing GitHub repository to it and how we solved that among other problems.
Continue reading →