{"?xml":{"@version":"1.0"},"edm:RDF":{"@xmlns:dc":"http://purl.org/dc/elements/1.1/","@xmlns:edm":"http://www.europeana.eu/schemas/edm/","@xmlns:wgs84_pos":"http://www.w3.org/2003/01/geo/wgs84_pos","@xmlns:foaf":"http://xmlns.com/foaf/0.1/","@xmlns:rdaGr2":"http://rdvocab.info/ElementsGr2","@xmlns:oai":"http://www.openarchives.org/OAI/2.0/","@xmlns:owl":"http://www.w3.org/2002/07/owl#","@xmlns:rdf":"http://www.w3.org/1999/02/22-rdf-syntax-ns#","@xmlns:ore":"http://www.openarchives.org/ore/terms/","@xmlns:skos":"http://www.w3.org/2004/02/skos/core#","@xmlns:dcterms":"http://purl.org/dc/terms/","edm:WebResource":[{"@rdf:about":"http://www.dlib.si/stream/URN:NBN:SI:doc-PJCF80WN/bb8de580-34af-40bc-9e17-2ac1232342ab/PDF","dcterms:extent":"574 KB"},{"@rdf:about":"http://www.dlib.si/stream/URN:NBN:SI:doc-PJCF80WN/57a216bb-c1de-478c-97d8-2cb5eb96d2c5/TEXT","dcterms:extent":"0 KB"}],"edm:TimeSpan":{"@rdf:about":"2006-2026","edm:begin":{"@xml:lang":"en","#text":"2006"},"edm:end":{"@xml:lang":"en","#text":"2026"}},"edm:ProvidedCHO":{"@rdf:about":"URN:NBN:SI:doc-PJCF80WN","dcterms:isPartOf":[{"@rdf:resource":"https://www.dlib.si/details/urn:nbn:si:spr-0y5dimiu"},{"@xml:lang":"sl","#text":"Elektrotehniški vestnik"}],"dcterms:issued":"2026","dc:creator":"Kukar, Matjaž","dc:format":[{"@xml:lang":"sl","#text":"številka:1-2"},{"@xml:lang":"sl","#text":"letnik:93"},{"@xml:lang":"sl","#text":"str. 14-31"}],"dc:identifier":["ISSN:0013-5852","COBISSID_HOST:280806915","URN:URN:NBN:SI:doc-PJCF80WN"],"dc:language":"sl","dc:publisher":{"@xml:lang":"sl","#text":"Elektrotehniška zveza Slovenije"},"dc:subject":[{"@xml:lang":"en","#text":"active feature acquisition"},{"@xml:lang":"sl","#text":"aktivno pridobivanje značilk"},{"@xml:lang":"en","#text":"machine learning"},{"@xml:lang":"sl","#text":"manjkajoči podatki"},{"@xml:lang":"en","#text":"missing data"},{"@xml:lang":"en","#text":"Shapley additive explanations"},{"@xml:lang":"sl","#text":"Shapleyeve razlage"},{"@xml:lang":"sl","#text":"strojno učenje"}],"dcterms:temporal":{"@rdf:resource":"2006-2026"},"dc:title":{"@xml:lang":"sl","#text":"Active feature acquisition by prediction explanations|"},"dc:description":[{"@xml:lang":"sl","#text":"In many real-world machine learning applications, particularly in resource-constrained domains such as medical diagnostics, acquiring feature values is a costly and often sequential process. Active Feature Acquisition (AFA) addresses this by selecting the most informative subset of features to acquire for a given instance, balancing predictive accuracy against acquisition costs. Conventional AFA strategies often rely on static, global feature importance metrics, which are instance-agnostic and can be suboptimal when feature relevance is context-dependent. We investigate a practically relevant two-stage acquisition scenario, where an initial subset of features is observed, and a subsequent non-overlapping subset is selected for acquisition. We propose a novel AFA strategy that leverages instance-specific model explanations, specifically Shapley additive explanations (SHAP), to guide the selection process. We conduct a systematic comparative study, evaluating three distinct acquisition strategies: random acquisition (as a baseline), acquisition guided by static global feature importance lists, and the proposed SHAP-based approach. Using XGBoost-trained models, these methods are evaluated across ten benchmark datasets under two missingness scenarios: Missing Completely at Random (MCAR) and Missing Not at Random (MNAR). The empirical results demonstrate that the SHAP-based strategy significantly outperforms static global feature importance methods in complex settings, particularly under MNAR missingness, where the context of observed features is critical for effective decision-making. While it also performs strongly in simpler MCAR scenarios (especially at high missingness rates), its robustness in more realistic settings suggests that utilizing instance-specific explanations provides a powerful and adaptive mechanism for personalized and effective feature acquisition"},{"@xml:lang":"sl","#text":"V mnogih prakticnih aplikacijah strojnega u ˇ cenja, npr. v ˇ medicinski diagnostiki, je pridobivanje dodatnih znacilk lahko ˇ drag in zamuden postopek. Aktivno pridobivanje znacilk ˇ (AFA) naslavlja ta problem, a enostavne strategije pogosto uporabljajo staticne, globalne metrike, ki so neodvisne od ˇ konteksta posameznega primera in zato suboptimalne. V clanku predlagamo novo strategijo, ki jo vodijo razlage ˇ SHAP, personalizirane za posamezen primer. Izvedli smo sistematicno primerjavo strategij za aktivno pridobivanje ˇ znacilk (naklju ˇ cna, stati ˇ cna in na razlagah osnovana) na ˇ desetih podatkovnih zbirkah, z mehanizmoma manjkajocih ˇ podatkov MCAR in MNAR. Empiricni rezultati ka ˇ zejo, da je ˇ strategija, osnovana na razlagah, znacilno bolj ˇ sa od stati ˇ cnih ˇ metod, zlasti v kompleksnih in realisticnih pogojih MNAR. ˇ Personalizirane razlage predstavljajo mocan in prilagodljiv ˇ nacin za u ˇ cinkovito pridobivanje dodatnih zna ˇ cilk"}],"edm:type":"TEXT","dc:type":[{"@xml:lang":"sl","#text":"znanstveno časopisje"},{"@xml:lang":"en","#text":"journals"},{"@rdf:resource":"http://www.wikidata.org/entity/Q361785"}]},"ore:Aggregation":{"@rdf:about":"http://www.dlib.si/?URN=URN:NBN:SI:doc-PJCF80WN","edm:aggregatedCHO":{"@rdf:resource":"URN:NBN:SI:doc-PJCF80WN"},"edm:isShownBy":{"@rdf:resource":"http://www.dlib.si/stream/URN:NBN:SI:doc-PJCF80WN/bb8de580-34af-40bc-9e17-2ac1232342ab/PDF"},"edm:rights":{"@rdf:resource":"http://rightsstatements.org/vocab/InC/1.0/"},"edm:provider":"Slovenian National E-content Aggregator","edm:intermediateProvider":{"@xml:lang":"en","#text":"National and University Library of Slovenia"},"edm:dataProvider":{"@xml:lang":"sl","#text":"Elektrotehniška zveza Slovenije"},"edm:object":{"@rdf:resource":"http://www.dlib.si/streamdb/URN:NBN:SI:doc-PJCF80WN/maxi/edm"},"edm:isShownAt":{"@rdf:resource":"http://www.dlib.si/details/URN:NBN:SI:doc-PJCF80WN"}}}}