다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

A model of learning from positive and negative examples is naturally described in terms of Formal Concept Analysis (FCA). In these terms, result of learning consists of two sets of intents (closed subsets of attributes): the first one contains intents that have only positive examples in the corresponding extents. The second one contains intents such that the corresponding extents contain only negative examples. On the one hand, we show how the means of FCA allows one to realize learning in this model with various data representation, from standard object-attribute one to that with labeled graphs. On the other hand, we use the language of FCA to give natural descriptions of some standard models of Machine Learning such as version spaces and decision trees. This allows one to compare several machine learning approaches, as well as to employ some standard techniques of FCA in the domain of machine learning. Algorithmic issues of learning with concept lattices are discussed. We consider applications of the concept-based learning, including Structure-Activity Relationship problem (in predictive toxicology) and spam filtering.

Machine Learning and Formal Concept Analysis

Today pesticides, antimicrobials and other pest control products used in conventional agriculture are questioned and alternative solutions are searched out. Scientific literature and local knowledge describe a significant number of active plant-based products used as bio-pesticides. The Knomana (KNOwledge MANAgement on pesticide plants in Africa) project aims to gather data about these bio-pesticides and implement methods to support the exploration of knowledge by the potential users (farmers, advisers, researchers, retailers, etc.). Considering the needs expressed by the domain experts, Formal Concept Analysis (FCA) appears as a suitable approach, due do its inherent qualities for structuring and classifying data through conceptual structures that provide a relevant support for data exploration. The Knomana data model used during the data collection is an entity-relationship model including both binary and ternary relationships between entities of different categories. This leads us to investigate the use of Relational Concept Analysis (RCA), a variant of FCA on these data. We consider two different encodings of the initial data model into sets of object-attribute contexts (one for each entity category) and object-object contexts (relationships between entity categories) that can be used as an input for RCA. These two encodings are studied both quantitatively (by examining the produced conceptual structures size) and qualitatively, through a simple, yet real, scenario given by a domain expert facing a pest infestation.

/pdf/effects-of-input-data-formalisation-in-relational-concept-3n67tqzu3v.pdf

Effects of Input Data Formalisation in Relational Concept Analysis for a Data Model with a Ternary Relation

https://hal-lirmm.ccsd.cnrs.fr/lirmm-02157786/document

FCA for software product line representation: Mixing configuration and feature relationships in a unique canonical representation

Formal Concept Analysis (FCA) and its associated conceptual structures are used to support exploratory search through conceptual navigation. Relational Concept Analysis (RCA) is an extension of Formal Concept Analysis to process relational datasets. RCA and its multiple interconnected structures represent good candidates to support exploratory search in relational datasets, as they are enabling navigation within a structure as well as between the connected structures. However, building the entire structures does not present an efficient solution to explore a small localised area of the dataset, to retrieve the closest alternatives to a given query. In these cases, generating only a concept and its neighbour concepts at each navigation step appears as a less costly alternative. In this paper, we propose an algorithm to compute a concept, and its neighbourhood, in connected concept lattices. The concepts are generated directly from the relational context family, and possess both formal and relational attributes. The algorithm takes into account two RCA scaling operators and it is implemented in the RCAExplore tool.

https://hal.archives-ouvertes.fr/hal-01735865/document

On-demand Relational Concept Analysis

Theoretical development and applications of graph signal processing (GSP) have attracted much attention. In classical GSP, the underlying structures are restricted in terms of dimensionality. A graph is a combinatorial object that models binary relations, and it does not directly model complex n-ary relations. One possible high dimensional generalization of graphs are simplicial complexes. They are a step between the constrained case of graphs and the general case of hypergraphs. In this paper, we develop a signal processing framework on simplicial complexes, such that we recover the traditional GSP theory when restricted to signals on graphs. It is worth mentioning that the framework works much more generally, though the focus of the paper is on simplicial complexes. We demonstrate how to perform signal processing with the framework using numerical examples.

Signal processing on simplicial complexes

A closed sets based learning classifier for implicit authentication in web browsing

Classification of sequential data (data obtained from series of actions in chronological order) has many applications in security, marketing or ergonomy. In this paper, we present a tool for classification of sequential data. We introduce a new clean dataset of web-browsing logs, and study the case of implicit authentification from web-browsing. We then detail more of the functioning of the tool and some of its parameters.

/pdf/a-tool-for-classification-of-sequential-data-36sn867dqy.pdf

A Tool for Classification of Sequential Data

Exploratory search allows to progressively discover a dataspace by browsing through a structured collection of documents. Concept lattices are graph structures which support exploratory search by conceptual navigation, i.e., navigating from concept to concept by selecting and deselecting descriptors. These methods are known to be limited by the size of concept lattices which can be too large to be efficiently computed or too complex to be browsed intelligibly. In this paper, we address the problem of providing techniques that reduce the complexity of FCA-based exploratory search. We show the suitability of AOC-posets, a condensed alternative structure to achieve conceptual navigation. Also, we outline algorithms to enable an on-demand generation of AOC-posets. The necessity to devise more flexible methods to perform product selection in software product line engineering is what motivates our work.

/pdf/on-demand-generation-of-aoc-posets-reducing-the-complexity-2pv7rl0t1d.pdf

Giacomo Kahn

Papers

Signal processing on simplicial complexes

A closed sets based learning classifier for implicit authentication in web browsing

On-demand Relational Concept Analysis

A Tool for Classification of Sequential Data

On-Demand Generation of AOC-Posets: Reducing the Complexity of Conceptual Navigation