Cognitive Computing

From knowledge graphs to cognitive computing

Share this post:

Editor’s note: This article originally appeared on 01net’s website as Dati di pubblica utilità: dai knowledge graph al cognitive computing; re-printed here with permission.

We live in the era of Big Data, and we use data for everything: from predicting which goods a customer may buy next, to forecasting the weather, and analyzing traffic in cities, or the spreading of diseases in the population of a country. Indeed data is an extremely powerful resource, an enabler for a large portion of our current technology.

However, we should not forget that though data is the foundation underpinning information processing, the real value is in the information that we are able to surface by processing the data, in the patterns that we are able to identify and recognize, and, ultimately, in the knowledge that we extract. So, we can say that from Big Data we usually distill Small Knowledge, which has Big Value.

The problem of representing knowledge in a machine processable format is a well known field of Artificial Intelligence, which has been extensively studied since the 1970s. An approach to knowledge representation based on graph (mathematical structure represented as sets of nodes, or vertices, which may be connected by edges) has been discussed for a long time, starting from the introduction of conceptual graphs [1] to the more recent Linked Data initiative [2] – a method to publish data and knowledge over the Web by explicitly representing their relationships, thus enabling computers to directly access and semantically query such distributed knowledge graphs (example below).

Knowledge graphs are a possible approach to building large knowledge bases, structured collections of facts about the world that computer systems can use to reason, and to interact with humans more naturally. These are some of the key characteristics of cognitive computing. The impact of this new computing approach are potentially very high, and, combined with other technologies current under development, it promises to open up whole new ways for humans to use computers.

Among the various fields of application of cognitive computing, a very interesting one is the integrated care domain, an emerging worldwide trend aiming at delivering more effective and coordinated forms of care provision spanning, among others, the social domain and the health domain. Social care and health care are knowledge intensive domain, where data and information are abundant. Delivering insights into the strengths and vulnerabilities of individuals with respect to the communities they live in and their social environment represents a key challenge for practitioners in the integrated care domain.

Link2Outcome [3] [4] is a prototype of a cognitive system that helps care workers to make more informed decision to improve outcome for patients. Among other goals, Link2Outcome aims at facilitating access to knowledge across domains and data sources, and to use such knowledge to summarize the state of an individuals though a dynamic set of vulnerability indexes. An interesting example of how Link2Outcome can help doctors is shown in [3]: the system may import (for example from an open data city portal) linked data describing pollution levels in a city, and correlate such information with the area where a patient lives in and with knowledge of how pollution may affect clinical conditions of the patient. A doctor may see how the vulnerability indexes of the patien change after the system has ingested the new data, and thus better understand how the environment may affect this patient.

As an extension to Link2Outcome, the more recent work BlueLENS [5], tackle the problem of collecting and providing the right information, to the right people, at the right time across the care continuum. The health of an individual is increasingly studied from multiple perspective, and analyzing multi-sectoral determinants of health is becoming a mainstream approach in integrated care. To enable this approach, data fusion from multiple heterogeneous sources is of key importance, and cognitive systems are uniquely positioned to help practitioners in surfacing the right information at the right time.

By establishing a more natural and seamless interaction between doctors and knowledge systems, cognitive computing is paving the way to revolutionize care, by bringing better and more timely insights that help experts in taking the best possible decisions to help patients.

L-R: Martin Stephenson​, Vanessa​ Lopez, Jiewen Wu,  Marco Luca Sbodio​,
Pierpaolo​ Tommasi, Spyros Kotoulas, Guruduth​ Banavar, and ​Nuno Lopes.

[1] John F. Sowa. 1976. Conceptual graphs for a data base interface. IBM Journal of Research and Development 20, 4 (July 1976), 336-357. DOI=[]

[2] Bizer, Christian; Heath, Tom; Berners-Lee, Tim (2009). “Linked Data—The Story So Far”. International Journal on Semantic Web and Information Systems 5 (3): 1–22. [doi:10.4018/jswis.2009081901]. ISSN 1552-6283.

[3] []

[4] Spyros Kotoulas, Vanessa Lopez, Marco Luca Sbodio, Martin Stephenson, Pierpaolo Tommasi, and Pol Mac Aonghusa. 2014. A linked data approach to care coordination. In Proceedings of the 25th ACM conference on Hypertext and social media (HT ’14). ACM, New York, NY, USA, 77-87. DOI=[]

[5] []

More stories

Dataset Lifecycle Framework: the swiss army knife for data source management in Kubernetes

Hybrid Cloud is rapidly becoming the go-to IT strategy for organizations seeking the perfect mix of scalability, performance and security. As a result, it is now common for an organization to rely on a mix of on-premise and cloud solutions, or “data-sources”, from different providers to store and manage their data. It doesn’t really sound […]

Continue reading

Archaeologists Seek to Unearth Mysterious Geoglyphs in Peru Using IBM AI and Geospatial Data

After uncovering a new Nasca Line formation with IBM Watson Machine Learning Accelerator on IBM Power Systems, Yamagata University will deploy IBM PAIRS in the hopes of further discoveries with AI.

Continue reading

Making Sense of Neural Architecture Search

It is no surprise that following the massive success of deep learning technology in solving complicated tasks, there is a growing demand for automated deep learning. Even though deep learning is a highly effective technology, there is a tremendous amount of human effort that goes into designing a deep learning algorithm.

Continue reading