arXiv subjects

Marc Jeanmougin

Publications and source records attributed to Marc Jeanmougin.

3 recordsLinked to original sources

arXiv2026

WiCleanData: Guaranteeing the Type Consistency of Wikidata by Taxonomy Refinement and Constraint Enforcement

Because of its collaborative nature, Wikidata suffers from errors, in- consistencies, and excessive complexity, such as redundant classes, ambiguity between instances and classes, wrong taxonomic paths, and type constraint violations. The manual curation of these issues is infeasible at scale. To address these challenges, we introduce WiCleanData, a refined version of Wikidata with a consistent tax- onomy and free from type constraint violations. Specifically, we have designed an automated pipeline that first cleans the taxonomy with language model assistance, then simplifies type constraints by hierarchical aggregation, and finally filters facts accordingly. The resulting knowledge graph, free from any type violation, is made publicly available via a Web interface, enabling easy exploration and downstream applications.

cs.AI↗

arXiv2023

A Self-Encoder for Learning Nearest Neighbors

We present the self-encoder, a neural network trained to guess the identity of each data sample. Despite its simplicity, it learns a very useful representation of data, in a self-supervised way. Specifically, the self-encoder learns to distribute the data samples in the embedding space so that they are linearly separable from one another. This induces a geometry where two samples are close in the embedding space when they are not easy to differentiate. The self-encoder can then be combined with a nearest-neighbor classifier or regressor for any subsequent supervised task. Unlike regular nearest neighbors, the predictions resulting from this encoding of data are invariant to any scaling of features, making any preprocessing like min-max scaling not necessary. The experiments show the efficiency of the approach, especially on heterogeneous data mixing numerical features and categorical features.

cs.LG↗

arXiv2013

A Proof of the Pumping Lemma for Context-Free Languages Through Pushdown Automata

The pumping lemma for context-free languages is a result about pushdown automata which is strikingly similar to the well-known pumping lemma for regular languages. However, though the lemma for regular languages is simply proved by using the pigeonhole principle on deterministic automata, the lemma for pushdown automata is proven through an equivalence with context-free languages and through the more powerful Ogden's lemma. We present here a proof of the pumping lemma for context-free languages which relies on pushdown automata instead of context-free grammars.

cs.FL↗