2026-04-22T00:35:04Zhttps://ebiltegia.mondragon.edu/oai/request
oai:ebiltegia.mondragon.edu:20.500.11984/59972024-03-06T07:50:53Zcom_20.500.11984_473col_20.500.11984_478
eBiltegia author Velez de Mendizabal, Iñaki author Ezpeleta, Enaitz author Zurutuza, Urko other Basto-Fernandes, Vitor other Méndez, José R. other Gomez-Meire, Silvana 2023-02-24T10:14:59Z 2023-02-24T10:14:59Z 2023 2376-5992 https://katalogoa.mondragon.edu/janium-bin/janium_login_opac.pl?find&ficha_no=171546 https://hdl.handle.net/20.500.11984/5997 Despite new developments in machine learning classification techniques, improving the accuracy of spam filtering is a difficult task due to linguistic phenomena that limit its effectiveness. In particular, we highlight polysemy, synonymy, the usage of hypernyms/hyponyms, and the presence of irrelevant/confusing words. These problems should be solved at the pre-processing stage to avoid using inconsistent information in the building of classification models. Previous studies have suggested that the use of synset-based representation strategies could be successfully used to solve synonymy and polysemy problems. Complementarily, it is possible to take advantage of hyponymy/hypernymy-based to implement dimensionality reduction strategies. These strategies could unify textual terms to model the intentions of the document without losing any information (e.g., bringing together the synsets “viagra”, “ciallis”, “levitra” and other representing similar drugs by using “virility drug” which is a hyponym for all of them). These feature reduction schemes are known as lossless strategies as the information is not removed but only generalised. However, in some types of text classification problems (such as spam filtering) it may not be worthwhile to keep all the information and let dimensionality reduction algorithms discard information that may be irrelevant or confusing. In this work, we are introducing the feature reduction as a multi-objective optimisation problem to be solved using a Multi-Objective Evolutionary Algorithm (MOEA). Our algorithm allows, with minor modifications, to implement lossless (using only semantic-based synset grouping), low-loss (discarding irrelevant information and using semantic-based synset grouping) or lossy (discarding only irrelevant information) strategies. The contribution of this study is two-fold: (i) to introduce different dimensionality reduction methods (lossless, low-loss and lossy) as an optimization problem that can be solved using MOEA and (ii) to provide an experimental comparison of lossless and low-loss schemes for text representation. The results obtained support the usefulness of the low-loss method to improve the efficiency of classifiers. eng © 2023 The Authors artificial intelligence Computational Linguistics Natural language processing Optimization Theory and Computation Spam filtering Synset-based representation Semantic-based feature reduction Multi-objective evolutionary algorithms Multi-objective evolutionary optimization for dimensionality reduction of texts represented by synsets http://purl.org/coar/resource_type/c_6501 
EBILTEGIA-N, MONDRAGON UNIBERTSITATEKO GORDAILU IREKIAN, MATERIALAK UZTEKO ERABILERA LIZENTZIA ETA ESKUBIDEEN LAGAPEN EZ-ESKLUSIBOA

Gordailuan utzitako materialaren egile edo egile eskubideen titular naizen aldetik, Mondragon Unibertsitateari erabilera-baimena eta/edo LAGAPEN EZ-ESKLUSIBOA ematen diot, doan, mundu eremuan eta legezko gehieneko epean, material horren gaineko erreprodukzio, banaketa, komunikazio publiko –elektronikoki eskuragarri jartzeko eskubidea barne– eta eraldatzeko eskubideetarako, hirugarren batzuen esku jartzeko aukera emanez, gordailuan utzitako material horri lotutako dagokion lizentzian ezarritako erabilera baldintzen arabera.

ONARTZEN DUT Mondragon Unibertsitateak, edukia aldatu gabe, gordailuan utzitako materiala gordetzeko beharrezkoak diren beste formatu edo bitarteko batzuetara eraldatzea edo egokitzea. Era berean, ONARTZEN DUT Mondragon Unibertsitateak gordailuan utzitako materialaren kopia bat baino gehiago gordetzea, segurtasun kopiak egiteko eta materiala etorkizunerako gordetzeko.

ZIURTATZEN DUT eta BERMATZEN DUT gordailuan utzitako materiala nik sortutako jatorrizko obra dela, eta nik dudala Lizentzia honetan jasotako lagapena emateko beharrezko egile eskubidearen titulartasuna.

Titulartasuna partekatzen den kasuetan, onartzen dut gainerako titularren baimena dudala Lizentzia hau emateko. Eskubideak hirugarren batzuei aurrez lagata izanez gero, ZIURTATZEN dut haien baimena dudala edo lizentzia honetan aurrez ikusitako moduan eskubideak erabiltzeko gaitasuna mantentzen dudala.

Lagatako materialak egile eskubideen titular ez naizen bestelako material batzuk izango balitu, ZIURTATZEN DUT eta BERMATZEN DUT dagokion titularraren baimena lortu dudala aurrez Lizentzia hau eman ahal izateko eta, era berean, ZIURTATZEN DUT eta BERMATZEN dut gordailuan utzitako materialaren edukian behar bezala identifikatuta eta aitortuta dagoela egile eskubidearen titular hori.

Azken batean, ZIURTATZEN DUT eta BERMATZEN DUT eBiltegian gordailuan utzitako materialak EZ DUELA HAUSTEN hirugarren batzuen eskubiderik, ez jabetza intelektualekorik ez industrialekorik, ohore, intimitate eta irudi eskubiderik, edo bestelako edozein eskubiderik.

Gordailuan utzitako materiala Mondragon Unibertsitatea ez den beste agentzia edo erakunderen batek finantzatutako ikerketan oinarrituta baldin badago, ZIURTATZEN DUT eta BERMATZEN DUT bete ditudala erakunde horrekin sinatutako kontratuak edo akordioak eskatutako betebehar guztiak.

Mondragon Unibertsitateak konpromisoa hartzen du gordailuan utzitako materialaren egile edo egile eskubideen titular gisa zu identifikatzeko, eta ez du material horretan Lizentzia honek espresuki ahalbidetutakoa ez den aldaketarik egingo.

Gordailuan utzitako materialaren egile edo egile eskubideen titular zaren heinean, obra eBiltegitik kentzeko eska dezakezu justifikatutako arrazoi batekin. Horretarako, arduradunekin harremanetan jarri beharko duzu.

Era berean, eBiltegiak gordailuan utzitako materiala kendu ahal izango du behar bezala justifikatutako kasuetan edo jabetza intelektualeko eskubideei lotuta hirugarren batzuen erreklamaziorik jasoz gero. Bi kasuetan aurretiaz espresuki jakinaraziko da.


-------------------------------------------------------


LICENCIA DE USO Y CESIÓN DE DERECHOS NO EXCLUSIVA PARA EL DEPÓSITO DE MATERIALES EN EBILTEGIA, REPOSITORIO ABIERTO DE MONDRAGON UNIBERTSITATEA

Como autor/a o titular de los derechos de autor del material depositado otorgo a Mondragon Unibertsitatea la autorización de uso y/o CESIÓN NO EXCLUSIVA, gratuita, con ámbito mundial y por el plazo máximo legal, de los derechos de reproducción, distribución, comunicación pública ‒incluido el derecho de puesta a disposición electrónica‒ y transformación sobre dicho material, permitiendo ponerlo a disposición de terceros, en las condiciones de uso que se establezcan en la correspondiente licencia asociada a dicho material depositado.

ACEPTO que Mondragon Unibertsitatea pueda, sin alterar el contenido, transformar o adaptar el material depositado a cualquier otro formato o medio que resulte necesario con fines de preservación. Igualmente, ACEPTO que Mondragon Unibertsitatea pueda guardar más de una copia del material depositado a fin de realizar copias de seguridad y preservar el material para el futuro.

ASEGURO y GARANTIZO que el material depositado es una obra original de mi creación, sobre la que ostento la titularidad de derechos de autor necesaria para otorgar la cesión contenida en la presente Licencia.

En caso de cotitularidad admito contar con el consentimiento de los restantes titulares para otorgar la presente Licencia. En caso de previa cesión de derechos a terceros, ASEGURO contar con su autorización o retengo la facultad de hacer uso de estos derechos en la forma prevista en la presente licencia.

En el caso de que el material depositado contuviera otro material del que no soy el/la titular de los derechos de autor, ASEGURO y GARANTIZO haber obtenido la previa autorización del correspondiente titular que me permite otorgar la presente Licencia y, del mismo modo, ASEGURO y GARANTIZO que en el contenido del material depositado queda correctamente identificado y reconocido el dicho titular de derecho de autor.

En definitiva, ASEGURO y GARANTIZO que el material depositado en eBiltegia NO INFRINGE derecho alguno de terceros, ya sea de propiedad intelectual o industrial, derecho al honor, a la intimidad y a la imagen, o cualquier otro derecho.

En el caso de que el material depositado esté basado en una investigación financiada por alguna agencia u organismo distinto de Mondragon Unibertsitatea, ASEGURO y GARANTIZO haber cumplido con todas las obligaciones requeridas por el contrato o acuerdo firmado con dicho organismo.

Mondragon Unibertsitatea adquiere el compromiso de identificarte como autor/a o titular de derechos de autor del material depositado, y no efectuará más alteraciones en dicho material que las expresamente permitidas en esta Licencia.

Como autor/a o titular de derechos de autor del material depositado puedes solicitar la retirada de la obra de eBiltegia por causa justificada. Para ello deberás ponerse en contacto con las personas responsables del mismo.

De igual modo, eBiltegia podrá retirar el material depositado en supuestos suficientemente justificados o en caso de reclamaciones de terceros relativas a los derechos de propiedad intelectual. En ambos supuestos existirá una notificación expresa previa.


-------------------------------------------------------


NON-EXCLUSIVE LICENCE FOR THE USE AND ASSIGNMENT OF RIGHTS FOR SUBMISSIONS TO EBILTEGIA, THE OPEN REPOSITORY OF MONDRAGON UNIBERTSITATEA

As the author or holder of the copyright of the sumitted material, I hereby grant Mondragon Unibertsitatea the NON-EXCLUSIVE authorisation to use and/or ASSIGN, free of charge, worldwide and for the maximum legal term, the rights to reproduce, distribute, publish ‒ including the right to make it available in electronic format ‒ and convert such material, allowing it to be made available to third parties, subject to the terms of use set forth in the respective licence associated with the submitted material.

I hereby AUTHORISE Mondragon Unibertsitatea to convert or adapt the submitted material to any other format or medium that may be necessary for the purpose of its preservation, without changing its content. I also AUTHORISE Mondragon Unibertsitatea to keep more than one copy of the submitted material so that it may make back-ups and preserve the material for the future.

I WARRANT and REPRESENT that the material submitted is an original work created by me, and I hold the necessary copyright to grant the right of assignment contained in this Licence.

In the event of joint ownership of author’s rights, I warrant that I have the consent of the other copyright holders to grant this Licence. In the event of any prior assignment of the rights to third parties, I WARRANT that I have their authorisation or still have the faculty to use these rights in the manner set forth in this licence.

If the submitted material contains any other material for which I do not hold the copyright, I WARRANT and REPRESENT that I have obtained the prior authorisation of the respective copyright holder, which enables me to grant this Licence and, similarly, I WARRANT and REPRESENT that in the content of the submitted material, the copyright holder is duly identified and acknowledged.

In short, I WARRANT AND REPRESENT that the material submitted to eBiltegia DOES NOT INFRINGE any third-party rights, whether relates to intellectual or industrial property, right to honour, privacy or self-image, or any other right.

If the submitted material is based upon research sponsored or supportes by any agency or institution other than Mondragon Unibertsitatea, I WARRANT AND REPRESENT that I have fulfilled all obligations required by the contract or agreement signed with the institution.

Mondragon Unibertsitatea hereby agrees to identify you as the author or holder of the copyright of the submited material, and it shall make no changes to that material other than those expressly permitted by this Licence.

As the author or holder of the copyright for the submitted material, you may request that the work is removed from eBiltegia if there is good cause. In such an event, you must contact the persons responsible for doing this.

Similarly, eBiltegia may remove the submitted material when there is good cause or in the event of third-party claims regarding intellectual property rights. In both cases, express prior notification shall be given.
 URL http://ebiltegia.mondragon.edu/xmlui/bitstream/20.500.11984/5997/1/Multi-objective%20evolutionary%20optimization%20for%20dimensionality%20reduction%20of%20texts%20represented%20by%20synsets.pdf File MD5 a7946aa2bd35ee3dba190eebb3571104 6489787 application/pdf Multi-objective evolutionary optimization for dimensionality reduction of texts represented by synsets.pdf URL http://ebiltegia.mondragon.edu/xmlui/bitstream/20.500.11984/5997/4/Multi-objective%20evolutionary%20optimization%20for%20dimensionality%20reduction%20of%20texts%20represented%20by%20synsets.pdf.txt File MD5 67a6d432f47db951719c96a2b156aca6 56876 text/plain Multi-objective evolutionary optimization for dimensionality reduction of texts represented by synsets.pdf.txt