Predicting Commodity Price Movements with Kontoor AI and WhereOS

Each day thousands of news articles are published by large news agencies. Reuters alone publishes over two million articles each year. On any given day, the world looks very confusing based on the news flow. Some companies are optimistic and increasing their investment, while some firms are going bankrupt.  Even for a professional analyst who has time and resources to devote for making sense of the economy, the job is overwhelming. Kontoor is a cloud based neural network that automatically reads in in real time each published news article published and creates an un-biased analysis of the news. The system stores all economic and corporate data from the news and makes sure all relevant information is captured from the news.

A news dashboard view and a predictive model developed with Kontoor and WhereOS

The challenge of automated data collection has traditionally been related to the fact that English language is extremely rich in vocabulary and expressions. Each event or phenomenon can be expressed in dozens of different ways, which makes traditional search engines ineffective. “Many platforms are capable of detecting synonyms for nouns and adjectives. Kontoor is the only platform capable of understanding the semantics for the verbs as well. Phrasal verbs are essential on English language. For instance, think of dozens of expressions there are for something increasing?”, says Jukka Taskinen, CEO of Kontoor.

“With Kontoor and WhereOS technologies, we’re able to engage the power of artificial intelligence or more precise advanced semantic neural networks extracting important business information from the news, and train machine learning algorithms e.g. to predict commodity price movements for different industry sectors”, says JP Partanen, CEO of WhereOS.

The collaboration between WhereOS and Kontoor leverages the strengths of each company: Kontoor’s ability to turn unstructured news into quantitative information, WhereOS’ abilities to integrate and operationalize multiple data source and develop machine learning models, API and applications on top of the underlying infrastructure.  “We have been extremely satisfied with WhereOS, as it has helped us to speed up our development of APIs and UIs based on our data with the factor or 5x”, says Jukka Taskinen, CEO of Kontoor. “It has been great to work with the truly talented team of Kontoor, and see how powerful Kontoor semantic artificial intelligence technology can really be in understanding and documenting the news feeds”, says JP Partanen, CEO of WhereOS.

Extracting Information from Unstructured News Text with Kontoor

When we are reading the news, we are looking for “events” – something that is taking place. This allows us to form an idea of the events that are shaping the direction of the world. Kontoor mimics this by creating generic events for companies and markets. Generic events cover any common activities by companies like increasing investment, hiring people, nominating a CEO, adjusting dividend or legal penalties. Also, events such as floods, strikes, and weather can be identified that have an impact on daily business. Events can be identified for companies, products and countries. Further, Kontoor identifies time periods, people and numeric data that will help in analysis.

During any news day, there is an endless stream of small stories about the companies you have never heard of and hence these stories are dismissed as insignificant. In other words, you are losing almost 100 percent of information in the news. All stories carry small pieces of data about how the business is performing and how they see the future. For instance, a story about Beiersdorf says they have increased sales and market share in their 3Q earning release.

“Beiersdorf, the maker of Nivea skin creams, said it snatched market share away from rivals as it posted a 6 percent rise in organic group sales in the first nine months.”

It is very difficult and labor intensive to read all the news and store it in a meaningful manner to be used later. Kontoor, on the other hand, is able to automatically collect and process such data about all events related for example related to selected corporations, and create a fully documented history for each firm.  You may not be interested in this company, but aggregating events data for the industry, it is possible to start understanding the big picture of the industry or product segment – trends become visible. By aggregating data for all industries, you will start seeing the trends in economy before the statistical organizations release data with a lag. Industry data can be divided into industry and services to predict PMI, industrial production and other similar indicators.

Kontoor Understands the Difference Between the Past and the Future

There is a significant difference in terms of economic value between the past events and the expected future events; Kontoor is specialized in understanding the difference of past and the future when processing the news. This enables Kontoor to gather automatically expectations related to commodity production changes, for instance, from the thousands of articles. Similarly, expectations for companies and markets can be gathered from the news. Therefore it is possible to find out a synthesis of opinions from the industry experts and companies

Let’s have a look at some practical examples: The airline company SAS expects its profits to improve as it reported 3Q earnings – a past event.  The steelmaker Thyssenkrupp sees its profitability to suffer. By gathering similar data for the industry, we can see whether these events are company specific or an indicator of an industry trend.

“Scandinavian airline SAS hiked its full-year earnings outlook on Friday as third-quarter profits topped market expectations.“

“Thyssenkrupp (TKAG.DE) cut the profit margin forecast for its capital goods business on Tuesday.”

Kontoor detects similarly expectations for sales, order, production, investment and so forth for any imaginable economic and corporate event.

Kontoor Semantic Artificial Intelligence Technology

Kontoor’s semantic deep neural network is trained by thousands and thousands of examples collected from the news; examples for each expression for corporate and economic events.  Based on the training material, network is being build up of multidimensional matrix of words that create clusters for each event we want to detect. Learning in the network happens as new expressions arrive that are close enough to some existing cluster. Network predicts the meaning of a new expression its proximity to exiting expressions.

Kontoor platform consist of core Kontoor technology and infrastructure around it – the infrastructure is similar to any search engine with artificial intelligence on top. Artificial intelligence consists of IBM Watson style, first generation component and a semantic neural network. Infrastructure is similar to any search engine to store indexed news stories.

Core technology is based on Stanford NLP library for natural language processing and different proprietary tools, models and databases, including machine learning model for sentiment analysis and deep neural network based models for events extraction. Training material for AI models constructed from huge unstructured news datasets, thousands of handcrafted rules and data engineering.

Infrastructure built around Kontoor technology consists of different Hadoop ecosystem technologies. Data storage organized with HDFS (Hadoop Distributed File System) and HBase. Data indexing and search features implemented with multi-server Solr Cloud installation and Cloudera HBase-Solr indexers. News tracking/fetching done by our own tracking module, streaming processing with Spark Streaming and Kafka messaging system. All together those technologies allow Kontoor fetch new data, process it, store and make results available with a minimal time delay. Kontoor platform is available via Play framework and ReactJS based user interface and Akka based API.

Creating Predictive Model and Application UI with WhereOS

WhereOS connects to Kontoor backend to receive a stream of quantitative data related to news (events, locations, time, sentiment, etc.). WhereOS can be used to create a layer of functionality on top of the data, by implemented different data processing pipelines in the WhereOS backend:

  1. Data APIs – The processed news data can be made available through APIs, and any new APIs create be dynamically created within minutes
  2. Combining the news data with other data sources – The quantitative data extracted from the news sources and be combined with other data sources, such as commodity price index information.
  3. Machine learning models – Data from the news can be used to train machine learning models (e.g. XGBoost in this case) against for example the commodity price index information (or any other quantitative data) and create ML model that can predict the e.g. the price index value based on processed news information feed received from Kontoor API.
  4. Application UI – Data can be easily visualised into different type of user interface applications for various purposes.

WhereOS uses Spark and Hive as the execution engine for the pipelines, which means the data processing is happening always as distributed computing, and can scale to billions of data points. WhereOS pipelines can be created using SQL and R programming languages including built-in functions for ETL (extract, transform, load) operations, statistical analysis, machine learning and artificial intelligence (AI), geographical and geospatial analysis, data visualization etc, which makes it really easy and simple to implement different data processing actions.

Jukka Taskinen, CEO of Kontoor:, +358 40 548 3780
More info:

JP Partanen, CEO & Chief Data Scientist of WhereOS:, +358 50 486 9257 Twitter LinkedIn
More info:

RoadCloud and WhereOS Collaborate on Automotive Data Collection and Processing

Traffic Jams in Helsinki

Are there traffic jams in Helsinki? People living in Helsinki would say YES, and people living in central Europe would probably laugh at this. The traffic jams are relatively mild compared to many other places but there are clearly times and locations where the traffic slows down significantly.

RoadCloud and WhereOS have joined forces to collect, analyze and visualize data collected from the fleet of commercial vehicles through RoadCloud sensors. “It’s amazing how detailed information you can collect from road network conditions through RoadCloud sensors”, says JP Partanen, CEO of WhereOS.

In the series of articles, we explain how the data can be collected from vehicle fleets and processed into models, that can further be used to solve different business problems such as traffic conditions, road conditions, or even creating machine learning / artificial intelligence model to predict these conditions based on other external variables such as weather forecasts. In this article, we explain how we created a video visualization of traffic conditions & jams throughout the day. “We are impressed with WhereOS, and how it quick it was to turn our data into an insightful video. We are integrating our data API to WhereOS and making our anonymized vehicle data easily accessible”, says Ari Tuononen, CEO of RoadCloud.

Capturing Vehicle Flow Data with RoadCloud IoT Sensors

RoadCloud has equipped commercial vehicle fleet with RoadCloud IoT sensors to collect and monitor vehicle data and road conditions. The sensors are automatically collecting basic information such speed, heading and acceleration, but more importantly information about road surface conditions such as road friction, road state (dry/snow/ice/water), temperature, bumps and pot holes as a few examples. The data collection is taking place 24/7, as the commercial vehicles are continuously throughout the day and night.

The sensor data is uploaded to the RoadCloud data backbone, where it is stored for further use & analysis. The data can be processed as historical data, or as a real-time feed of continuous updating data.

Analyzing and Visualizing the Sensor Data with WhereOS

WhereOS connects to the RoadCload data backbone and can process the data further. In this visualization, the RoadCloud data is processed and combined with OpenStreetMap (OSM) street network data in several pipelines:

  1. OSM data preparation pipeline extracts the OpenStreetMap street network for the desired region, Helsinki capital area in this case, and splits the streets into segments of desired length (e.g. 200m).
  2. RoadCloud data extraction pipeline loads data from RoadCloud data backbone, and assigns joins the GPS (latitude & longitude) data points into corresponding street segments. Joining uses WhereOS geokey/geohash based operation for matching massive amount of geographical shapes – GPS points and polylines/linestrings (street segments) – together.
  3. NTILE pipeline takes the joined data, and approximates maximum speed for each street segment, by taking average of the 10% highest speeds driven on that specific street segment.
  4. Hourly traffic pipeline calculates average speed driven for each individual street segment for each hour, and also the speed decrease from the maximum speed.
  5. Rendering pipeline takes speed decrease for each street segment for each hour and produces an MP4 video where each frame represents traffic conditions (speed decrease) for each street segment for the given hour. Each street segment is colored so that red equals to high speed decrease (i.e. high traffic / traffic jam) and green means no decrease (i.e. traffic speed close to maximum).

WhereOS uses Spark and Hive as the execution engine for the pipelines. WhereOS pipelines can be created using SQL and R programming languages including built-in functions for ETL (extract, transform, load) operations, statistical analysis, machine learning and artificial intelligence (AI), geographical and geospatial analysis, data visualization etc.

Traffic Jam Video & Further Innovation

The data used for this video has been collected from the vehicles over the time of one full year (2018). You can see how the rush hour traffic affects the average speeds at different locations on different times of the day.

The RoadCloud data can be used for many other analyses such as: How the road conditions – friction, road state (dry/wet/ice/snow), temperature – affect the average speeds. Or how a speed bump or some other new traffic arrangement affects the traffic around it. In the upcoming articles we will dig deeper into the data and create new interesting visualizations out of it.

Ari Tuononen, CEO, RoadCloud:, +358 50 5604 702
More info: Twitter LinkedIn

JP Partanen, CEO, WhereOS:, +358 50 486 9257
More info: Twitter LinkedIn

Most Extensive Visualization of US Income: Over 33000 Zip Codes Visualized

This application visualizes mean income of households in United States, for each zip code (ZCTA, zip code tabulation area): Click here to open the application

Background: Urban Decay

The visualization shows clearly, how large cities in US are surrounded by wealthy zip code areas (red rings). Inner cities, however are areas with less income (green or yellow), with few exceptions, such as New York, Washinton DC and San Francisco.

“In the United States during the 1940s, for the first time a powerful interaction between segregation laws and race differences in terms of socioeconomic status enabled white families to abandon inner cities in favor of suburban living. The result was severe urban decay that, by the 1960s, resulted in crumbling “ghettos“. Prior to national data available in the 1950 US census, a migration pattern of disproportionate numbers of whites moving from cities to suburban communities was easily dismissed as merely anecdotal. Because American urban populations were still substantially growing, a relative decrease in one racial or ethnic component eluded scientific proof to the satisfaction of policy makers. In essence, data on urban population change had not been separated into what are now familiarly identified its ‘components.’ The first data set potentially capable of proving ‘white flight’ was the 1950 census.” Source: Wikipedia

The visualization is based on American Community Survey (ACS) from

Red = high income
Yellow = medium income
Green = low income


I’m looking for bloggers, who want to co-author articles about interesting/surprising facts about US (or European) population demographics, economy etc. along with visualizations. Also, if you have ideas for further visualizations, please contact me via

If you want to embed this app to your web page, read the instructions here.


Finland: 58 Interesting Indicators Visualized

This is an app, through which you can visualize demographic data from Finland. Try also clicking for example “Relative” and “Higher level university degree”, or “Employed”.

Click here to open the application

The data used in this visualization: Paavo – Open data by postal code area, Statistics Finland. The material was downloaded from Statistics Finland’s interface service with the licence CC BY 4.0.

Check this article how to embed a WhereOS app to your own website. If you have ideas what kind of data you’d like to see in the apps, write me an email


How to Embed a WhereOS App

Embedding a WhereOS apps to any web page or a blog is easy. Just include the an HTML iframe-tag with width and height properties, and point the link to the application you have created.

URL Format:


[domain] is the name of the service domain you have used when starting the WhereOS service for yourself

[application_name] is the name of the application

[ui_view_name] is the name of the UI view, if more than one UI view exists for the given application

Remeber to replace spaces in the application name with underscore in the URL. Application with name “Reachability Analysis” becomes Reachability_Analysis and so forth.

<iframe frameborder="0" style="width: 100%; height: 400px;" src=""></iframe>


WhereOS – A new way of doing Spark, Hive and SQL

WhereOS is a new cloud based operating system for distributed, data driven applications, that’s built on top of Apache Spark. WhereOS makes it easier to build, host, distribute and share applications built using (big) data.   

WhereOS can be programmed through SQL programming language including built-in functions for ETL (extract, transform, load) operations, statistical analysis, geographical and spatial analysis, data visualization etc. Programs written in WhereOS applications are always executed as distributed processes, without developers explicitly design their applications for parallel processing and scalability.

WhereOS uses Spark and Hive as the execution engine for the applications. WhereOS can be extended through two types of drivers: function and data asset drivers. Function drivers can introduce any new Hive/SparkSQL functions, to implement new features for the SQL language. Data asset drivers provide new ways of integrating data formats and/or data transmission protocols into the system.

If you want to discuss more, contact us through


Location Analytics Industry Trends

Here are three relevant industry trends, why I believe WhereOS is going to change the world:

  1. Hyperconnected World – Businesses are getting more and more connected to online, cloud based information systems as a part of their business processes. Customers are accessing companies’ online services 24/7 through different interfaces, generating more and more information about their purchase behavior, location, movement, social interactions and so forth – all tied to geographical locations
  2. Business Digitalization – New businesses are challenging traditional industries by using combinations of static and dynamic information to create new services for their digital customers: They know better who their customers are, what they want, where they are located, and how to serve them best. They know where the operational bottlenecks are and can deliver the right goods and services to right locations to create the best customer experience, maximize the sales and minimize the cost.
  3. Data Explosion – The amount of data businesses are collecting from their customers and business operations is increasing constantly. Data sets are large, and in constant change – the existing IT systems and processes do not facilitate innovation in the fast paced change many industries are undergoing.

With WhereOS, it’s extremely easy to built APIs and applications based on data: instead of months, new applications can be create in a matter of hours or days. This facilitates lean prototyping of how data can be used to create new value-adding services and products, helping both traditional and new businesses to speed up the development.

If you are interested in discussing more, please contact me through


Route Optimization

Route Optimization application calculates the most efficient route that goes through the specified set of waypoints. Route optimization can be used to calculate service routes, delivery routes, pick-up routes etc., that require all waypoints to be visited in the most efficient manner.

The most efficient route is defined calculated through R function, that takes the characteristics of the route and waypoints into account. The default weighting function is duration based, i.e. the optimal route is the one that finishes earliest, with two waypoint categories: critical and low. Critical waypoints are visited first, and low category waypoints last, in a manner that the shortest duration route is the final result. The R function can be modified to take into account different route characteristics, such as route distance, waypoint types, waypoint SLAs, time of day etc.

If you have ideas or questions – or you need help to use the application in your business domain, write to us

Open the app: