Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Semantics Overview

At present, only the Overview and Opportunities sections are finished. This presentation will be updated as the work progresses. The outline for the unfinished sections are in place for your reference. The PDF password is "Saraswati".

Joel Natividad

on 11 May 2012

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Semantics Overview

Semantics??? Overview Problem Statement Current Approach The Semantic Way Data from Scarce to Superabundant
Social Multiplier
Everyone produces Data
Data Underutilized
Data is Siloed From "Needle in a Haystack" To "Needles in a Field of Haystacks" Traditional BI Shortcomings Still building plumbing
Yet another SILO
Can only answer questions that
were conceived ahead of time
ETL woes - loading, massaging, transform
Not real-time
Cannot cope up with the Data Tsunami

"Dumb" - BI systems cannot infer nor reason The Main Difference Inference & Reasoning From Searching To Knowing - Spectrum of Knowledge Representation and Reasoning Capabilities "Smart Data" XML RDF Ontologies Inference Rules Semantics! <Yorkie>Coco</Yorkie> (Joel) (is_dogowner_of) (Coco) mammal -> dog -> Yorkie -> Coco (Maya) (is_spouse_of) (Joel) If (personA) (is_spouse_of) (personB), then (personA) (is_dogowner_of) (dog) + + + = "Obvious" relationships are not understood by the computer
Each "fact" has to be explicitly declared
The system does not "discover" logical facts & relationships
Certain concepts/relationships cannot be easily expressed
"Closed World" - the system does not know anything beyond its boundaries Searching instead of Knowing We still have to MANUALLY collate the data Employee Hours Wasted per Task Hierarchies of concepts, like: Relations, in triples, like: Customized tags, like: Semantics 101 <Yorkie>Coco</Yorkie> INFERS <dog>Coco</dog>, <mammal>Coco</mammal>

(Maya) (Is_spouse_of) (Joel) INFERS (Maya) (is_dogowner_of) (Coco) We infer/discover additional Facts & Relationships The Coming Semantic Wave
A Tidal Wave of Four Internet Growth Stages Web 1.0 - Static Web (1991-2003)

Web 2.0 - Social Networks (2004-2010)

Web 3.0 - Semantic Web (starting NOW)

Web 4.0 - The Ubiquitous Web (future-2020?) Opportunities Knowledge is encapsulated in Opaque Software
Data Schemas reflect limited Knowledge
Data Organization is tightly coupled with the Schema
Schemas support limited Data Integrity
Very hard to Localize Data - gather, collect, and concentrate data from different data sources
Same source data Localized redundantly by many systems Limitations of Conventional
Application Development Knowledge is represented by an Ontology
Data Organization is decoupled from the Schema
Inferencing creates new Knowledge
Defines the Meaning of Data
Enables data integration across heteregenous silos
Utilizes "reasoners" to ensure Data Integrity
All Semantic data can be web addressable
SOA "out-of-the-box"
Ontology can change without altering underlying facts/assertions allowing rapid response to business change Benefits of Semantic
Application Development * Economist Special Report Attached The Coming
Data Tsunami DROWNING IN DATA Here's how we extract Knowledge
using the Traditional Approach Semantics is not only for Business Intelligence
We can fix the Knowledge Gap at the Source Conventional
Semantic <rdf:RDF xmlsns="http://natividads.com/metadata/mammal"> The Bottom Line "Smart" Data/Applications Knowledge is Baked-in
New Knowledge can be inferred
Agility to Adopt to Ever-changing Conditions
Semi-automatic Data Integration
Machine Intelligence Increased Business Agility
New Business Capabilities & Opportunities
Dramatically cut back IT expenditures
While Building, Leveraging & Preserving your current IT investments
Is INEVITABLE In Plain English Where most systems are Conventional
SOA Semantics Over the next decade, Web 3.0 will spawn multi-billion dollar technology markets that will drive trillion dollar global economic expansions to transform industries as well as our experience of the internet. This BI industry is estimated to be worth more than $100 billion and growing at almost 10% a year, roughly twice as fast as the software business as a whole. Mills Davis, Semantic Wave Report 2008, Project 10X Data, Data Everywhere
The Economist, Feb 2010 Today’s business intelligence (BI) and reporting
systems are NOT designed for this on-the-fly creation of meaning. These systems lack the capability to capture and manage the semantics of the business in a more dynamic, scalable way. Systems that are able to “understand” and have far greater contextual awareness will provide a level of proactive assistance that was previously available only from human helpers. For scientists, this will mean deeper scientific insight, richer discovery, and faster breakthroughs. PWC Tech Forecast, Spring 2009 Craig Mundie, Microsoft Chief Research & Strategy Officer
The Fourth Paradigm, Data-Intensive Scientific Discovery
Microsoft Research 2009 What do the Experts Say? The real reason why the semantic web is inevitable at this point is because the business potential of a web of data — inside the enterprise and out in the public web — is becoming obvious and immense. Businesses are data-driven. Scott Brinker, President & CTO,
Ion Interactive, SemTech 2009
June 2009 The Semantic Web isn't just inevitable.
It is imminent. Semantic Technology Products & Services Opportunities Top 10 Opportunities in the Enterprise 1 Information sharing

2 Semantic search, discovery, & navigation

3 Semantic mashups and composite applications

4 Semantic infrastructure / middleware SSOA, SBPM, SWS, virtualization, policy-based computing

5 Semantic business intelligence

6 Semantic ERP applications CRM, PLM, SCM, HRM

7 Semantic governance, compliance, & risk

8 Semantic web sites, wikis, collaboration, interest networking, & collective knowledge systems

9 Semantic advertising, marketing, personalization, & customization

10 Intelligent systems knowledge-based research, design, engineering, simulation, planning, scheduling, optimization, & decision support. Mills Davis, Web 3.0 Manifesto, Project 10X, Oct 2008 2010 - Total BI Market estimated at 100 Billion
Semantic Market estimated at 10 Billion By 2020, Mills Davis estimates
that the Entire BI Market will be Semantic 2015 -BI Market Size - 175 Billion
Semantic BI - 52 Billion Source: Mills Davis, Project 10X
PWC Spring 2009 Tech Forecast And that is just the Business Intelligence Market,
within the Next Decade, the majority of Information and Communication Technologies (ICT) will have Semantic Capabilities.

It is Inevitable. There is no other way ICT can cope up with
the Data Tsunami. Semantic Wave The Semantic Web can assist the evolution of human knowledge as a whole. Tim Berners-Lee, "Inventor" of the Internet
Scientific American, 2001 (1991-2003) (2004-2010) (NOW-???) (The FUTURE) Be sure to read the Semantic Briefing Portfolio at http://bit.ly/J2yInC PDF password (case-sensitive): "Saraswati" Semantics!!! =
Knowledge Engineering Knowledge Engineering
Practice NOW kNOW kNOW 2 Organizations need to make quick decisions using all the information that they have

IT should enable this but often can’t deliver quickly enough

Analysis requires us to complete all these steps with our information:
Create it—works great
Store it—works great
Change it—works great
Find it—not so good
Integrate it—very complex and difficult
Analyze it—finally!
Use it—great if you’re using the app that created it; otherwise a BIG problem

How can we analyze all our information across all platforms,
wherever it sits, quickly?
Full transcript