quinta-feira, 10 de setembro de 2015

[Ontobras 2015] Keynote: Developing Ontologies for Land Cover and Land Use Data - Gilberto Câmara

Motivation
Seeking to connect the knowledge from different fields. E.g. economy, geography etc. How to connect such knowledge? How to understand the cultural aspects of each field? This requires a huge effort and it is a challenge in the Ontology field.

Quest for the perfect map:
The Ontologist (Barry Smith like) point of view:
The world is divided in cells. Each cell has a single class. There is a correct classification. The more our classification coincides with the ideal classification, the better.

But...
It is very hard to use this point of view in practice. The map's cells have a specific granularity, which depends on the instrument used to measure the world (satellite, eye, or another measurement instrument). And given this granularity, it is impossible to provide a unique classification to each cell.

Land use: the arrangements, activities, inputs people undertake in a certain land cover type to produce, change or maintain it.
Classification: non-managed forest, cattle production, temporary agriculture, shifting cultivation.



The definition depends on a debate among real people. In this case, the speaker was on a helicopter with Marina Silva, the governor of Amazon, military personnel and other people. Each one was using a particular definition of how many land should be covered to characterise a forest (and differentiate it from deforested area).

There is something missing on the Ontological debate: change, time, movement
Representing change is very hard!

Land trajectories: the transformation of land cover due to actions of land use. He showed graphs showing that the same area was a forest from 2000 to 2004, a pasture from 2004 to 2006 ...

E.g. of Land trajectory graph and the relation to event


How does our brain represent time?
... by means of events: relevant moments of change
Book: "Why time speeds up when you get older" - dutch author

Objects exist and events occur (*it sounds like the distinction between endurant and event).

The fact that our brain is not capable of processing change in a good way (we develop different senses whether there was a big or small change, our perceptions change with time etc.), it makes it really hard to perceive time (which affects the perception of land trajectory). And it is even harder to describe with a unique category that would cover all parts of the trajectory.

Land cover vs. Land use
Land cover - endurant (*according to Giancarlo, it is the role of the land).
Land use - event.

Opportunity: because now (and even more in the future), we have much more data to represent the states of the world, we are more able to understand land trajectory. The challenge is to make sense of the data and for that, ontological thinking is essencial.

*brilliant slide!


"In theory, there is no difference between theory and practice. In practice, there is." (Yogi Berra)

Conclusions:
Managing change is a major challenge for the scientific community. Big data creates new challenges. We need ontological thinking for understanding data.

quarta-feira, 9 de setembro de 2015

[Ontobras 2015] Keynote: Large-scale Semantic Web Reasoning - by Grigoris Antoniou

Large-scale Semantic Web Reasoning - by Grigoris Antoniou

1st Keynote Ontobras

Intro: Big Data is commonly associated with Data Mining and Machine learning.
·      Uncover hidden patterns, thus new insights
·      Mostly statistical approaches

He claims semantics and reasonings are also relevant:
1. Semantic interoperability
2. Decision making
3. Data cleaning
4. Inferring high-level knowledge from data

1. Why do we need Semantic Interoperability?
  • ·      To create added value through combination of different, independently maintained sources. E.g. Health (combine healthcare, social and economic data to better predict problems and to derive interventions)
  • ·      Combine historic and actual data
Problems: complexity and dynamicity

2. Make sense of the huge amounts of data:
  • -       turn it into action
  • -       be able to explain decisions – transparency and increased confidence
  • -       be able to deal with imperfect, missing or conflicting data
  • -       all in the remit of Knowledge Representation (KR)
  • -       E.g. alert of a possible dangerous situation for an elderly person when certain conditions are met.
Potential Domains: smart cities, intelligent environments, ambient assisted living, intelligent healthcare (including remote monitoring), disaster management

But can we deliver?

The problem:
  • -       traditional approaches work in centralized memory
  • -       But we cannot load big data (or the Web) on a centralized memory, nor are we expected to do so in the future.
  • To the rescue: New computational paradigms
  • -       Developed in the past decade as part of high-performance computing, cloud computing etc.
  • -       Developed independently of the Semantic Web (SW) and KR, but we can use them.
What Follows:
  •  Basic RDFS reasoning on Map Reduce
  •  Computationally simple nonmonotonic reasoning on Map Reduce
  •  Computationally complex ontology repair approach using Signal/Collect (can we apply these approaches to exponentially reasoning tasks?)
Problems:
  • Load Balancing 
  • High I/O Cost 
  • Program Complexity
MapReduce:
  • Introduced by Google in 2004
  • Computation is expressed only with Maps and Reduce (popular implementation: Hadoop)
No restante da palestra, ele propos soluções utilizando MapReduce.

Still missing / Future Work:
- What pararelization architecture are more appropriate?
- There are no agreed benchmarks for large-scale reasoning
- More complex types of reasoning should be considered, such as spationtemporal reasoning and exponential reasoning.
- Extreme reasoning: make reasoning with big data work in real time (there are new systems, like Apache Storm)