Blog – Domingo López Rodríguez

Categories

All (23)

Algebra (1)

Algorithm (7)

Algorithms (1)

Axioms (1)

Bioinformatics (1)

Bonds (1)

Combinatorial optimization (2)

Conversational AI (1)

Cybersecurity (2)

Data Analysis (1)

Data analysis (2)

Data mining (3)

Divide-and-conquer (1)

E-health (1)

E-learning (2)

Energy efficiency (1)

Engineering (2)

FCA (12)

Formal concept analysis (1)

Fuzzy logic (1)

Fuzzy-FCA (7)

GIS (1)

GPU (1)

Implications (1)

Knowledge representation (1)

Knowledge space (1)

Lattice theory (1)

Logic (2)

MRI (1)

Machine learning (2)

Malware (2)

NLP (1)

Network Coding (1)

Neural networks (3)

Neuroimage (1)

Performance (2)

Physics (1)

Principal component analysis (1)

Publication idea (1)

R (1)

R package (1)

Recommendation system (1)

Recommender systems (2)

Remote Sensing (1)

Sentiment Analysis (1)

Simplification (1)

Software (3)

Sustainability (2)

Text Mining (1)

Theoretical CS (5)

Thermodynamics (1)

Time series (1)

Uncertainty (1)

Unsupervised learning (1)

Urban Planning (1)

Urban planning (1)

fcaR (2)

Analyzing fuzzy data doesn’t have to be slow: we taught FCA algorithms to ‘speak fuzzy’ natively

FCA

Fuzzy-FCA

Algorithms

Performance

fcaR

Analyzing data with graded values (fuzzy) has always been a bottleneck. We present a new family of native algorithms that are dramatically faster than traditional ‘scaling’ methods.

Domingo López-Rodríguez, Manuel Ojeda-Hernández, Ángel Mora, Carlos Bejines

Divide and conquer… the ‘if-then’ rules in your data

FCA

Algorithm

Implications

Divide-and-conquer

Knowledge representation

The ‘CARVE’ algorithm is a super-fast ‘divide-and-conquer’ method for finding data concepts. We upgraded it to ‘CARVE+’ so it can find all the ‘if-then’ logical rules at the same time, giving us the best of both worlds.

Domingo López-Rodríguez, Manuel Ojeda-Hernández, Tim Pattison

We built an open-source tool to find a city’s worst ‘heat island’ hotspots

Software

Sustainability

Data mining

Urban planning

GIS

We present URSUS_UHI, a new open-source software tool that helps urban planners automatically find the most unfavourable areas in a city: the places with the highest temperatures and the least green space.

Francisco Rodríguez-Gómez, José del Campo-Ávila, Domingo López-Rodríguez, Luis Pérez-Urrestarazu

Beyond just ‘correct’: new logic rules to make data insights easier to read

FCA

Logic

Theoretical CS

Algorithm

Simplification

Our previous work created a logic to find correct rules from positive and negative data. Now, we’ve added new logical equivalences to ‘shrink’ those rules, making them shorter, simpler, and more useful for humans.

Domingo López-Rodríguez, Manuel Ojeda-Hernández, Carlos Bejines

Real-world data is vague: a new method for analyzing time series with fuzzy numbers

Fuzzy logic

Time series

Data analysis

Algorithm

Uncertainty

A stock price isn’t just a point, it has volatility. We introduce a new method to convert standard time series into ‘fuzzy time series’ to better model and analyze this real-world uncertainty.

Hugo J. Bello, Manuel Ojeda-Hernández, Domingo López-Rodríguez, Carlos Bejines

Malware names are a mess: we used fca to build a ‘family tree’ for threats

FCA

Cybersecurity

Malware

Data mining

Different antivirus tools call the same threat different names. We used Formal Concept Analysis (FCA) to automatically build a unified, logical hierarchy from this chaos, making malware easier to track.

Manuel Ojeda-Hernández, Domingo López-Rodríguez, Ángel Mora

Malware names are a mess: we used fca to build a ‘family tree’ for threats

FCA

Cybersecurity

Malware

Data mining

Different antivirus tools call the same threat different names. We used Formal Concept Analysis (FCA) to automatically build a unified, logical hierarchy from this chaos, making malware easier to track.

Manuel Ojeda-Hernández, Domingo López-Rodríguez, Ángel Mora

Invited seminar Formal Concept Analysis with R

Formal concept analysis

R package

Recommendation system

The second week of March, two of our team members were invited to give a seminar about the use of the R package fcaR at the University of Cádiz in Spain.

Domingo López Rodríguez, Ángel Mora Bonilla

A new language for network codes: using fuzzy vector spaces to understand flags

Fuzzy-FCA

Theoretical CS

Algebra

Network Coding

Flag codes are a powerful tool for network data transfer. We show how a more abstract language, fuzzy vector spaces, provides a new and deeper way to understand their fundamental structure.

Carlos Bejines, Manuel Ojeda Hernández, Domingo López-Rodríguez

How a 100-year-old math quirk helps us clean up mri brain scans

Neuroimage

Machine learning

Data analysis

MRI

Axioms

We found that noiseless MRIs follow Benford’s Law (the ‘1s’ rule). We use this discovery to build a new machine learning tool that can accurately measure the noise level in brain scans.

Rosa Maza-Quiroga, Karl Thurnhofer-Hemsi, Domingo López-Rodríguez, Ezequiel López-Rubio

Beyond a single dataset: connecting knowledge with ‘bonds’

FCA

Fuzzy-FCA

Bonds

Theoretical CS

Recommender systems

How do you logically link ‘users’ to ‘movie genres’? We explore the theory of ‘bonds’, the mathematical ‘glue’ that connects two different datasets, providing new properties and experiments to make them more powerful.

Ondrej Krídlo, Domingo López-Rodríguez, Lubomir Antoni, Peter Eliaš, Stanislav Krajči, Manuel Ojeda-Aciego

Beyond ‘is’ or ‘is not’: a fuzzy algebra for ‘almost-concepts’

FCA

Fuzzy-FCA

Theoretical CS

Lattice theory

In FCA, a pair is either a concept or it’s not. We built a ‘graded’ algebraic structure that includes ‘almost-concepts’ (preconcepts, semiconcepts) and reveals a much richer hierarchy.

Manuel Ojeda-Hernández, Domingo López-Rodríguez, Pablo Cordero

Ditch the generic dictionary: building smarter sentiment analysis with fca

FCA

Sentiment Analysis

Text Mining

NLP

Algorithm

Standard sentiment dictionaries often fail. We present a new method using Formal Concept Analysis (FCA) to automatically create custom dictionaries from your data, which achieves better performance.

Manuel Ojeda-Hernández, Domingo López-Rodríguez, Ángel Mora

Finding our cities’ hotspots: using satellites to identify areas that are too hot and have no green space

Sustainability

Remote Sensing

Data Analysis

Urban Planning

Urban heat islands are a major health risk. We developed an automatic method using satellite imagery to find the ‘most unfavourable’ areas (high heat, low vegetation) across 16 major Spanish cities.

Francisco Rodríguez-Gómez, Rafael Fernández-Cañero, Gabriel Pérez, José del Campo-Ávila, Domingo López-Rodríguez, Luis Pérez-Urrestarazu

Introducing fcaR: the first complete formal concept analysis package for R

FCA

R

Software

Fuzzy-FCA

fcaR

R is the top language for data science, but it lacked a key tool. We built ‘fcaR’, the first comprehensive R package to bring the power of logic-based data mining (FCA) to the R community.

Pablo Cordero, Manuel Enciso, Domingo López-Rodríguez, Ángel Mora

What students know (and what they don’t): building a ‘knowledge space’ with fca

FCA

E-learning

Knowledge space

Algorithm

Publication idea

We use fca with both positive and negative attributes (e.g., ‘student failed this’) to analyze a math course. This new method finds the ‘key’ items that define the course’s true knowledge structure.

Manuel Ojeda-Hernández, Ángel Mora, Francisco Pérez-Gámez, Domingo López-Rodríguez, Nicolás Madrid

Logic isn’t just about what’s there: simplifying rules with positive and negative data

FCA

Logic

Theoretical CS

Algorithm

Fuzzy-FCA

In data, knowing ‘what is absent’ is as important as ‘what is present’. We developed a new ‘mixed simplification logic’ to clean and remove redundancy from rules that use both positive and negative attributes.

Francisco Pérez-Gámez, Domingo López-Rodríguez, Pablo Cordero, Ángel Mora, Manuel Ojeda-Aciego

Building a ‘smart’ medical chatbot: a conversational recommender for diagnosis using fuzzy rules

FCA

Fuzzy-FCA

Recommender systems

Conversational AI

E-health

We designed a conversational AI for medical diagnosis that doesn’t just match keywords. It uses fuzzy logic and formal concept analysis to have an intelligent dialogue, understand uncertain symptoms, and guide a user to a diagnosis.

Pablo Cordero, Manuel Enciso, Domingo López-Rodríguez, Ángel Mora

Beyond just speed: how to be energy-efficient with gpus in bioinformatics

GPU

Bioinformatics

Performance

Energy efficiency

GPUs are fast, but they are power-hungry. We analyzed exactly when and how to get the most ‘performance per watt’ for key biomedical applications, and the answer isn’t just ‘buy a bigger gpu’.

Juan Pérez, Andrés Rodríguez, Juan Francisco Chico, Domingo López-Rodríguez, Manuel Ujaldón

Taking it to the next level: a new ‘multivalued’ neural net to identify mechanical designs

Neural networks

Combinatorial optimization

Engineering

Algorithm

We designed an advanced ‘multivalued’ neural network to solve the classic kinematic chain isomorphism problem. It converges rapidly, is fully automatic, and requires no parameter tuning, outperforming previous approaches.

Gloria Galán Marín, Domingo López-Rodríguez, Enrique Mérida Casermeiro

Giving neural maps a new dimension: a probabilistic pca self-organizing map

Neural networks

Principal component analysis

Machine learning

Unsupervised learning

We present a new neural model that extends the classic self-organizing map (SOM) by placing a probabilistic principal component analyzer (PPCA) at each neuron. The result is a powerful tool for visualizing high-dimensional data.

Ezequiel López-Rubio, Juan Miguel Ortiz-de-Lazcano-Lobato, Domingo López-Rodríguez

A new neural network for a classic engineering problem: identifying identical mechanisms

Neural networks

Combinatorial optimization

Engineering

How can you tell if two complex mechanical designs are secretly the same? We taught a Hopfield neural network to solve this classic engineering problem automatically and efficiently.

Gloria Gálan-Marín, Enrique Mérida-Casermeiro, Domingo López-Rodríguez

Can a ‘demon’ teach thermodynamics? our pedagogical simulation of maxwell’s paradox

E-learning

Software

Physics

Thermodynamics

We built a computer simulation to help students intuitively understand one of the most famous and challenging thought experiments in physics: maxwell’s demon.

Domingo López, Carlos Criado