/
Tomás Pariente Lobo – Atos Spain Tomás Pariente Lobo – Atos Spain

Tomás Pariente Lobo – Atos Spain - PowerPoint Presentation

kittie-lecroy
kittie-lecroy . @kittie-lecroy
Follow
412 views
Uploaded On 2017-01-29

Tomás Pariente Lobo – Atos Spain - PPT Presentation

Final Review Meeting Luxembourg 20 November 2012 Conclusion Overall Project Objectives Technical Impact Near realtime and scalable financial unstructured data acquisition extraction analysis integration and visualization ID: 515140

increase financial analysis visualization financial increase visualization analysis time real project markets sentiment information achievements models based providers knowledge

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "Tomás Pariente Lobo – Atos Spain" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

Slide1

Tomás Pariente Lobo – Atos SpainFinal Review Meeting, Luxembourg. 20 November 2012

ConclusionSlide2

Overall Project Objectives

Technical

Impact

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization pipelineWell grounded, deep semantic sentiment analysis (using financial and knowledge representation theories)Advanced financial decision support models based on high-level featuresAdvanced real-time visualization techniquesReal-world use cases using the underlying FIRST infrastructure

Open source and commercial versions to maximize business possibilities.Increase the ability to exploit very large financial information spaces for citizens and professionals.Decrease information asymmetry and increase market transparency at the financial markets and thus contribute to making such markets more trustworthy.Increase the competitiveness of European financial service providers and institutions

2Slide3

Project Achievements

Technical

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization

pipelineWell grounded, deep semantic sentiment analysis (using financial and knowledge representation theories)Advanced financial decision support models based on high-level features

Advanced real-time visualization techniquesReal-world use cases using the underlying FIRST infrastructure

3

3,159 RSS feeds

15 million unique documents collectedSlide4

Project Achievements

Technical

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization

pipelineWell grounded, deep semantic sentiment analysis (using financial and knowledge representation theories)Advanced financial decision support models based on high-level featuresAdvanced real-time visualization techniquesReal-world use cases using the underlying FIRST infrastructure

5 degrees of positive & negative sentimentLabeled by domain uc# experts

Knowledge-based approach

4Slide5

Project Achievements

Technical

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization

pipelineWell grounded, deep semantic sentiment analysis (using financial and knowledge representation theories)Advanced financial decision support models based on high-level featuresAdvanced real-time visualization techniquesReal-world use cases using the underlying FIRST infrastructure

PriceVolaP&L

Assesshighlowgood unaccmedhighgood

unacc

med

Med

good

acc

low

Low

acc

acc

low

Low

good

good

5

Quantitative models for

doc. categorization

,

P&D detection

, and Twitter sentiment classification

Qualitative models for RIM and P&DSlide6

Project Achievements

Technical

Gauge

Canyon

Confidence

Sentiment

with volume

6

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization

pipeline

Well grounded, deep

semantic sentiment analysis

(using financial and knowledge representation theories)

Advanced

financial decision support models

based on high-level features

Advanced

real-time visualization

techniques

Real-world

use cases

using the underlying FIRST infrastructureSlide7

Project Achievements

Technical

Near real-time

and scalable financial

unstructured data acquisition, extraction, analysis, integration, and visualization

pipelineWell grounded, deep semantic sentiment analysis (using financial and knowledge representation theories)Advanced financial decision support models based on high-level featuresAdvanced real-time visualization techniquesReal-world use cases using the underlying FIRST infrastructure

7

Qualitative

Quantitative

VisualizationSlide8

Project Achievements

Impact

Open source and commercial versions

to maximize business possibilities.

Increase the ability to

exploit very large financial information spaces for citizens and professionals.Decrease information asymmetry and increase market transparency at the financial markets and thus contribute to making such markets more trustworthy.Increase the competitiveness of European financial service providers and institutions

8

Proprietary

https://github.com/project-first

External usage

/ commercialSlide9

Project Achievements

Impact

Open source and commercial versions

to maximize business possibilities.

Increase the ability to

exploit very large financial information spaces for citizens and professionals.Decrease information asymmetry and increase market transparency at the financial markets and thus contribute to making such markets more trustworthy.Increase the competitiveness of European financial service providers and institutions

9Slide10

Project Achievements

Impact

Open source and commercial versions

to maximize business possibilities.

Increase the ability to

exploit very large financial information spaces for citizens and professionals.Decrease information asymmetry and increase market transparency at the financial markets and thus contribute to making such markets more trustworthy.Increase the competitiveness of European financial service providers and institutions

10

Document

sentences

Objects

Features

Sentiment

Drill down

X

The glassbox modelSlide11

Project Achievements

Impact

Open source and commercial versions

to maximize business possibilities.

Increase the ability to

exploit very large financial information spaces for citizens and professionals.Decrease information asymmetry and increase market transparency at the financial markets and thus contribute to making such markets more trustworthy.Increase the competitiveness of European financial service providers and institutions

11

Competitive advantage for financial services providers

Competitive advantage for usecase providers Slide12
Slide13

Thanks

13