This document proposes using a learning classifier system (LCS) to mine association rules from data streams online by adapting to changes in associations. It describes CSar, an LCS that represents rules with intervals for continuous attributes and uses antecedent or consequent grouping to organize rules into association sets. Experimental results show CSar can extract rules with similar support and confidence as Apriori on discrete data and quantitative association rules on continuous data, though further analysis is needed to compare CSar to other rule miners and test its ability to adapt to changing data streams.
Deep neural networks have revolutionized the data analytics scene by improving results in several and diverse benchmarks with the same recipe: learning feature representations from data. These achievements have raised the interest across multiple scientific fields, especially in those where large amounts of data and computation are available. This change of paradigm in data analytics has several ethical and economic implications that are driving large investments, political debates and sounding press coverage under the generic label of artificial intelligence (AI). This talk will present the fundamentals of deep learning through the classic example of image classification, and point at how the same principal has been adopted for several tasks. Finally, some of the forthcoming potentials and risks for AI will be pointed.
Deep neural networks have boosted the convergence of multimedia data analytics in a unified framework shared by practitioners in natural language, vision and speech. Image captioning, lip reading or video sonorization are some of the first applications of a new and exciting field of research exploiting the generalization properties of deep neural representation. This tutorial will firstly review the basic neural architectures to encode and decode vision, text and audio, to later review the those models that have successfully translated information across modalities. The contents of this tutorial are available at: https://telecombcn-dl.github.io/2019-mmm-tutorial/.
Deep neural networks have revolutionized the data analytics scene by improving results in several and diverse benchmarks with the same recipe: learning feature representations from data. These achievements have raised the interest across multiple scientific fields, especially in those where large amounts of data and computation are available. This change of paradigm in data analytics has several ethical and economic implications that are driving large investments, political debates and sounding press coverage under the generic label of artificial intelligence (AI). This talk will present the fundamentals of deep learning through the classic example of image classification, and point at how the same principal has been adopted for several tasks. Finally, some of the forthcoming potentials and risks for AI will be pointed.
Deep neural networks have boosted the convergence of multimedia data analytics in a unified framework shared by practitioners in natural language, vision and speech. Image captioning, lip reading or video sonorization are some of the first applications of a new and exciting field of research exploiting the generalization properties of deep neural representation. This tutorial will firstly review the basic neural architectures to encode and decode vision, text and audio, to later review the those models that have successfully translated information across modalities. The contents of this tutorial are available at: https://telecombcn-dl.github.io/2019-mmm-tutorial/.
Do not Match, Inherit: Fitness Surrogates for Genetics-Based Machine Learning...Xavier Llorà
A byproduct benefit of using probabilistic model-building genetic algorithms is the creation of cheap and accurate surrogate models. Learning classifier systems---and genetics-based machine learning in general---can greatly benefit from such surrogates which may replace the costly matching procedure of a rule against large data sets. In this paper we investigate the accuracy of such surrogate fitness functions when coupled with the probabilistic models evolved by the x-ary extended compact classifier system (xeCCS). To achieve such a goal, we show the need that the probabilistic models should be able to represent all the accurate basis functions required for creating an accurate surrogate. We also introduce a procedure to transform populations of rules based into dependency structure matrices (DSMs) which allows building accurate models of overlapping building blocks---a necessary condition to accurately estimate the fitness of the evolved rules.
Linkage Learning for Pittsburgh LCS: Making Problems TractableXavier Llorà
Presentation by Xavier Llorà, Kumara Sastry, & David E. Goldberg showing how linkage learning is possible on Pittsburgh style learning classifier systems
https://telecombcn-dl.github.io/2018-dlcv/
Deep learning technologies are at the core of the current revolution in artificial intelligence for multimedia data analysis. The convergence of large-scale annotated datasets and affordable GPU hardware has allowed the training of neural networks for data analysis tasks which were previously addressed with hand-crafted features. Architectures such as convolutional neural networks, recurrent neural networks and Q-nets for reinforcement learning have shaped a brand new scenario in signal processing. This course will cover the basic principles and applications of deep learning to computer vision problems, such as image classification, object detection or image captioning.
MOMENT: Temporal Meta-Fact Generation and Propagation in Knowledge GraphsParis Sud University
MOMENT: Temporal Meta-Fact Generation and Propagation in Knowledge Graphs [F. Saïs, J. E. Gonzales Malaverri and G. Quercini @ACM-SAC-SWA 2020]
This paper deals with the problem of temporal meta-fact genera- tion in RDF knowledge graphs (KGs). These temporal meta-facts represent the time validity of facts, for instance, <Barack Obama, presidentOf, United States of America> is valid for the period [2008..2016]. We propose an approach called MOMENT that com- bines two methods, the first uses a set of specified rules to generate meta-facts in knowledge bases where no temporal meta-fact exist. The second method exploits existing temporal meta-facts and a set of Horn rules generated by AMIE [9] to propagate the meta-facts and thus expand the set of temporal meta-facts. An experimental evaluation has been conducted using Yago, DBpedia and Wikidata datasets. The obtained results are promising and showed the relevance of such an approach for temporal meta-fact generation in Knowledge Graphs.
An Ontology-based Decision Support Framework for Personalized Quality of Life...Marina Riga
Publication:
Riga, M., Kontopoulos, E., Karatzas, K., Vrochidis, S., Kompatsiaris, I. (2018) An Ontology-based Decision Support Framework for Personalized Quality of Life Recommendations. In Dargam F. et al. (Eds.) Proceedings of the 4th International Conference on Decision Support System Technology (ICDSST 2018), LNBIP 313, pp. 38–51, Heraklion, Greece, 22–25 May 2018, doi:10.1007/978-3-319-90315-6_4
Model Attribute Check Company Auto PropertyCeline George
In Odoo, the multi-company feature allows you to manage multiple companies within a single Odoo database instance. Each company can have its own configurations while still sharing common resources such as products, customers, and suppliers.
How to Make a Field invisible in Odoo 17Celine George
It is possible to hide or invisible some fields in odoo. Commonly using “invisible” attribute in the field definition to invisible the fields. This slide will show how to make a field invisible in odoo 17.
Read| The latest issue of The Challenger is here! We are thrilled to announce that our school paper has qualified for the NATIONAL SCHOOLS PRESS CONFERENCE (NSPC) 2024. Thank you for your unwavering support and trust. Dive into the stories that made us stand out!
Unit 8 - Information and Communication Technology (Paper I).pdfThiyagu K
This slides describes the basic concepts of ICT, basics of Email, Emerging Technology and Digital Initiatives in Education. This presentations aligns with the UGC Paper I syllabus.
Welcome to TechSoup New Member Orientation and Q&A (May 2024).pdfTechSoup
In this webinar you will learn how your organization can access TechSoup's wide variety of product discount and donation programs. From hardware to software, we'll give you a tour of the tools available to help your nonprofit with productivity, collaboration, financial management, donor tracking, security, and more.
A Strategic Approach: GenAI in EducationPeter Windle
Artificial Intelligence (AI) technologies such as Generative AI, Image Generators and Large Language Models have had a dramatic impact on teaching, learning and assessment over the past 18 months. The most immediate threat AI posed was to Academic Integrity with Higher Education Institutes (HEIs) focusing their efforts on combating the use of GenAI in assessment. Guidelines were developed for staff and students, policies put in place too. Innovative educators have forged paths in the use of Generative AI for teaching, learning and assessments leading to pockets of transformation springing up across HEIs, often with little or no top-down guidance, support or direction.
This Gasta posits a strategic approach to integrating AI into HEIs to prepare staff, students and the curriculum for an evolving world and workplace. We will highlight the advantages of working with these technologies beyond the realm of teaching, learning and assessment by considering prompt engineering skills, industry impact, curriculum changes, and the need for staff upskilling. In contrast, not engaging strategically with Generative AI poses risks, including falling behind peers, missed opportunities and failing to ensure our graduates remain employable. The rapid evolution of AI technologies necessitates a proactive and strategic approach if we are to remain relevant.
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
The Roman Empire A Historical Colossus.pdfkaushalkr1407
The Roman Empire, a vast and enduring power, stands as one of history's most remarkable civilizations, leaving an indelible imprint on the world. It emerged from the Roman Republic, transitioning into an imperial powerhouse under the leadership of Augustus Caesar in 27 BCE. This transformation marked the beginning of an era defined by unprecedented territorial expansion, architectural marvels, and profound cultural influence.
The empire's roots lie in the city of Rome, founded, according to legend, by Romulus in 753 BCE. Over centuries, Rome evolved from a small settlement to a formidable republic, characterized by a complex political system with elected officials and checks on power. However, internal strife, class conflicts, and military ambitions paved the way for the end of the Republic. Julius Caesar’s dictatorship and subsequent assassination in 44 BCE created a power vacuum, leading to a civil war. Octavian, later Augustus, emerged victorious, heralding the Roman Empire’s birth.
Under Augustus, the empire experienced the Pax Romana, a 200-year period of relative peace and stability. Augustus reformed the military, established efficient administrative systems, and initiated grand construction projects. The empire's borders expanded, encompassing territories from Britain to Egypt and from Spain to the Euphrates. Roman legions, renowned for their discipline and engineering prowess, secured and maintained these vast territories, building roads, fortifications, and cities that facilitated control and integration.
The Roman Empire’s society was hierarchical, with a rigid class system. At the top were the patricians, wealthy elites who held significant political power. Below them were the plebeians, free citizens with limited political influence, and the vast numbers of slaves who formed the backbone of the economy. The family unit was central, governed by the paterfamilias, the male head who held absolute authority.
Culturally, the Romans were eclectic, absorbing and adapting elements from the civilizations they encountered, particularly the Greeks. Roman art, literature, and philosophy reflected this synthesis, creating a rich cultural tapestry. Latin, the Roman language, became the lingua franca of the Western world, influencing numerous modern languages.
Roman architecture and engineering achievements were monumental. They perfected the arch, vault, and dome, constructing enduring structures like the Colosseum, Pantheon, and aqueducts. These engineering marvels not only showcased Roman ingenuity but also served practical purposes, from public entertainment to water supply.
IWLCS'2008: First Approach toward Online Evolution of Association Rules with Learning Classifier Systems
1. First Approach toward Online
Evolution of Association Rules
with Learning Classifier Systems
Albert Orriols-Puig1
Orriols Puig
Jorge Casillas2
Ester Bernadó-Mansilla1
1Research Group in Intelligent Systems
Enginyeria i Arquitectura La Salle, Ramon Llull University
2Dept. Computer Science and Artificial Intelligence
University of Granada
2. Framework
Michigan-style LCSs are reaching maturity
– First successful implementations (Wil
(Wilson, 1995; Wil
1995 Wilson, 1998)
• Many other derivations YCS, UCS, XCSF, and many others
– Applications in important domains
• Data mining (Bernadó et al, 02; Wilson, 02; Bacardit & Butz, 04; Butz, 06;
Orriols t l
O i l et al, 2008)
• Function approximation (Wilson, 2002; Butz et al. 2008)
• Reinforcement Learning (Lanzi et al, 2006, Butz et al. 2006)
• Clustering (Tamee et al., 2006; 2007)
– Theoretical analyses for design (Butz et al., 2004; 2007; Drugowitsch, 2008)
Enginyeria i Arquitectura la Salle Slide 2
GRSI
3. Motivation
Which types of problems have we shown to be able to solve?
– Data mining (supervised learning)
g( p g)
• Classification
• Function approximation
– Reinforcement learning
–D
Data streaming: H area i the ML community
i Hot in h i
• Learn online from stream of examples
– Supervised learning
– Unsupervised learning
• Two edited books last year: (Aggarwal, 2007; Gama, 2007)
• Decision tree able to learn online in JMRL: (Nuñez et. al, 2007)
Enginyeria i Arquitectura la Salle Slide 3
GRSI
4. Aim
General purpose
Mine association rules online from streams of examples,
adapting the model to association drift
d ti th d lt i ti d ift
Why?
– Many industrial processes generate streams unlabeled data
• Marketing problem
• Modeling Spanish power consumption
– LCS seems to provide a natural support to mine data online
– There are not many approaches to this problem
Particular scope of the present work?
First proposal of an architecture able to mine
association rules from streams of data
i ti lf t fd t
Enginyeria i Arquitectura la Salle Slide 4
GRSI
5. Outline
1. Mining Association Rules
1 Mi i A ii Rl
2.
2 Description of CSar
3. Experimental Methodology
4. Results
5. Conclusions and Further Work
Enginyeria i Arquitectura la Salle Slide 5
GRSI
6. Mining Association Rules
g
Goal: Extract l th t d
G l E t t rules that denote relationships among variables
t l ti hi i bl
with high support and confidence from data sets
Market basket transactions (Data set) Rule set with rules IF X THEN Y
N. Items
IF diapers THEN beer
1 diapers, bread, ham, cheese, jam, beer
IF bread and ham THEN cheese
2 beer, diapers, monkfish,
beer diapers monkfish sausage
3 salmon, beer, bread, diapers, ham, cheese
…
...
Rules evaluation: support and confidence
pp
Enginyeria i Arquitectura la Salle Slide 6
GRSI
7. Approaches to Mine Association Rules
pp
First approaches on categorical data
– AIS (A
(Agrawal, 1993)
l
Many of them, two-phase alg.:
– Apriori (Agrawal & Srikant, 1994)
1. Obtain frequent item sets
q
– Many others 2. Create rules from these item sets
y g
– Limitation: Only categorical data
Quantitative Association Rules Mining
– Di
Discretize and apply A i i (Srikant & Agrawal, 1996; Fukuda et al, 1996)
ti d l Apriori
– Interval-based Association Rules (Mata et al., 2002)
– Fuzzy association rules (Yager, 1995; Hong, 2001)
Online association rule mining (Wang et al., 2007)
al
Enginyeria i Arquitectura la Salle Slide 7
GRSI
8. Outline
1. Mining Association Rules
1 Mi i A ii Rl
Rule Representation
2.
2 Description of CSar
Process Organization
3. Experimental Methodology
4. Results
5. Conclusions and Further Work
Enginyeria i Arquitectura la Salle Slide 8
GRSI
9. Description of CSar
p
Rule representation
IF xi is vi and … and xj is vj THEN xk is vk
– Problems with ℓ attributes
– Antecedent formed by a conjunction of ℓa variables (0 < ℓa < ℓ)
– Consequent contains a single variable
– Each variable is defined by
• Nominal attributes A single value
• Continuous attributes An interval of maximum size [ℓi, ui];
[ ]
» Restriction: ui - ℓi < maxInt
– Example:
IF age in [20,25] and height in [5,6] and sport=yes THEN weight is [160, 170]
Slide 9
GRSI
10. Description of CSar
p
Classifier Parameters
1.
1 Support (supp): occurring frequency of the rule
2. Confidence (
2 C fid (conf): strength of the i li i
f) h f h implication
3. Fitness (F): computed from support and confidence
4. Association set size ( ) average size of the association sets th rule
4A i ti t i (as): i f th i ti t the l
participates in
5 Numerosity (n): copies of the rule
5.
6. Experience (exp): number of examples matched by the rule antecedent
Enginyeria i Arquitectura la Salle Slide 10
GRSI
11. Description of CSar
p
Stream of
Environment examples
Match Set
M t h S t [M]
1A C sup conf F n as exp
Covering
3A C sup conf F n as exp
Population [P] if |[M]| < θmna
5A C sup conf F n as exp
6A C sup conf F n as exp
…
1A C sup conf F n as exp
2A C sup conf F n as exp
Generation of
3A C sup conf F n as exp
all possible
4A C sup conf F n as exp
association sets [A]
5A C sup conf F n as exp
Match set
6A C sup conf F n as exp
generation
…
Each [A] is given a selection
p
probability proportional to
yp p
its average confidence
Classifier
Parameters
Update [A] selection
Deletion
Association Set [A]
Selection, Reproduction,
mutation
Genetic Algorithm Run [A]
[]
3 A C sup conf F n cs as exp
p p
subsumption
6 A C sup conf F n cs as exp
…
Enginyeria i Arquitectura la Salle Slide 11
GRSI
12. Description of CSar
p
Creation of association set candidates
Match Set [M]
[A] 1
1A C sup conf F n as exp
[A] 2
3A C sup conf F n as exp
3 A C sup conf F n cs as exp
5A C sup conf F
p n as exp
p
6 A C s p conf F
sup n cs as e p
exp
6A C sup conf F n as exp 3A C sup conf F n cs as exp
…
… 6A C sup conf F n cs as exp
…
Goal:
Organize the classifiers of [M] in different [A]
Similar classifiers should participate in the same [A]
Strategies:
Antecedent grouping
Consequent grouping
Enginyeria i Arquitectura la Salle Slide 12
GRSI
13. Description of CSar
p
Antecedent grouping:
– Put in [A] the rules with the same variables in the antecedent
– Underlying idea: Rules with the same antecedent may express similar
conclusions
Consequent grouping:
– For nominal rep., put in [A] rules with same consequent and values
– F interval-based rep., put overlapped rules i th same [A]
For i t lb d t l d l in the
– Underlying idea: Rules that describe the same range of values of the
same concept may have similar antecedents
concept, antecedents.
Enginyeria i Arquitectura la Salle Slide 13
GRSI
14. Outline
1. Mining Association Rules
1 Mi i A ii Rl
2.
2 Description of CSar
3. Experimental Methodology
4. Results
5. Conclusions and Further Work
Enginyeria i Arquitectura la Salle Slide 14
GRSI
15. Experimental Methodology
p gy
First experiment:
– Compare CSar with Apriori (Agrawal & Srikant, 1994)
– D t set with nominal values: th zoo problem
Data t ith ill the bl
• 15 binary attributes and 2 categorical attributes
Second experiment
S d i
– Analyze CSar in problems with real-valued attributes
– The Wisconsin breast-cancer diagnosis problem
• 9 continuous attributes and 1 nominal attribute
CSar configuration
Iterations = 100,000, popSize = 6,400, θGA = 50, v = 10, θdel = 10 …
Evaluation method: Count the number of rules for a certain support and
confidence
Enginyeria i Arquitectura la Salle Slide 15
GRSI
16. Outline
1. Mining Association Rules
1 Mi i A ii Rl
2.
2 Description of CSar
3. Experimental Methodology
4. Results
5. Conclusions and Further Work
Enginyeria i Arquitectura la Salle Slide 16
GRSI
17. First experiment: the Zoo Problem
p
Antecedent grouping Consequent grouping
• Antecedent grouping maintains more niches and so creates more rules
• Rules with high confidence and support are approximately the same in both cases
Enginyeria i Arquitectura la Salle Slide 17
GRSI
18. First Experiment: the Zoo Problem
p
The three learners discover the same number of rules of high confidence and support.
As the support and confidence decreases
A-priori
A i i creates a larger number of rules th C
t l b f l than Csar
CSar with antecedent grouping evolves more rules than Csar with consequent
grouping
Enginyeria i Arquitectura la Salle Slide 18
GRSI
19. Second Experiment: Continuous Data
p
We want to analyze:
– Whether CSar can extract a population of interval based rules
interval-based
with high support and confidence
– Differences between consequent and antecedent grouping
– The effect of varying the maximum size allowed for an interval
Apriori not included since it only deals with categorical data
Which experiments?
– Move maximum interval to {0.1, 0.9}
– Use the same configuration as in experiment 1
Enginyeria i Arquitectura la Salle Slide 19
GRSI
20. Second Experiment: Continuous Data
p
MAXIMUM INTERVAL SIZE = 0.1
01
Antecedent grouping Consequent grouping
• Antecedent grouping maintains more niches and so creates more rules
• Rules with high confidence and support are approximately the same in both cases
Enginyeria i Arquitectura la Salle Slide 20
GRSI
21. Second Experiment: Continuous Data
p
MAXIMUM INTERVAL SIZE = 0.9
09
Antecedent grouping Consequent grouping
• As the maximum size of the interval increases both configurations can learn rules
with more support and confidence
• The number of rules in the consequent grouping reduces considerably, since
subsumption applies and subsumes rules in favor of the most general ones
Enginyeria i Arquitectura la Salle Slide 21
GRSI
22. Outline
1. Mining Association Rules
1 Mi i A ii Rl
2.
2 Description of CSar
3. Experimental Methodology
4. Results
5. Conclusions and Further Work
Enginyeria i Arquitectura la Salle Slide 22
GRSI
23. Conclusions
Michigan-style LCS can extract association rules
– Similar number of rules with high support and confidence to this of
Apriori
– Besides, CSar can extract quantitative association rules
g
Niching is crucial to evolve different association rules
– Antecedent grouping yield more rules when support and
confidence are low than consequent g p g
q grouping
– Both groupings resulted in a similar number of rules with high
support and confidence
pp
But still much needs to be analyzed
Enginyeria i Arquitectura la Salle Slide 23
GRSI
24. Further Work
Compare CSar to other quantitative association rule miners
Studies on the population size required to sustain associations
Add fuzzy representation
– Deal more effectively with the problem of maximum interval size
Mining changing environments
– Can CSar adapt quickly to association changes?
– Can we mine large data streams with fixed population size and
get the most relevant associations
Enginyeria i Arquitectura la Salle Slide 24
GRSI
25. First Approach toward On-line
Evolution of Association Rules
with Learning Classifier Systems
Albert Orriols-Puig1
Orriols Puig
Jorge Casillas2
Ester Bernadó-Mansilla1
1Research Group in Intelligent Systems
Enginyeria i Arquitectura La Salle, Ramon Llull University
2Dept. Computer Science and Artificial Intelligence
University of Granada