{"id":"https://openalex.org/W7225028353","doi":null,"title":"Evaluating chunking strategies across document types and their impact on RAG pipeline and AIagent performance","display_name":"Evaluating chunking strategies across document types and their impact on RAG pipeline and AIagent performance","publication_year":2026,"publication_date":"2026-10-28","ids":{"openalex":"https://openalex.org/W7225028353"},"language":"en","primary_location":{"id":"pmh:oai:repozitorij.upr.si:IzpisGradiva.php?id=23967","doi":null,"is_oa":true,"landing_page_url":"https://repozitorij.upr.si/IzpisGradiva.php?id=23967","pdf_url":"https://repozitorij.upr.si/Dokument.php?id=35569&dn=","source":{"id":"https://openalex.org/S4377196486","display_name":"Repository of the University of Primorsk (University of Primorska)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I118905719","host_organization_name":"University of Primorska","host_organization_lineage":["https://openalex.org/I118905719"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/bachelorThesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://repozitorij.upr.si/Dokument.php?id=35569&dn=","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160263230","display_name":"Ilija Petrov","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Petrov, Ilija","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5160263230"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9106000065803528,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9106000065803528,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.07909999787807465,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.001500000013038516,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/document-chunking","display_name":"document chunking","score":0.8270000219345093},{"id":"https://openalex.org/keywords/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.925000011920929},{"id":"https://openalex.org/keywords/chunking-strategies","display_name":"chunking strategies","score":0.7070000171661377},{"id":"https://openalex.org/keywords/scientific-articles","display_name":"scientific articles","score":0.7149999737739563},{"id":"https://openalex.org/keywords/contracts","display_name":"contracts","score":0.8550000190734863},{"id":"https://openalex.org/keywords/parliamentary-proceedings","display_name":"parliamentary proceedings","score":0.3160000145435333},{"id":"https://openalex.org/keywords/web-articles","display_name":"web articles","score":0.8999999761581421},{"id":"https://openalex.org/keywords/raga","display_name":"raga","score":0.6209999918937683},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.5410000085830688},{"id":"https://openalex.org/keywords/ai-agents","display_name":"AI agents","score":0.6639999747276306},{"id":"https://openalex.org/keywords/vector-embeddings","display_name":"vector embeddings","score":0.6949999928474426},{"id":"https://openalex.org/keywords/question-answering","display_name":"question answering","score":0.3869999945163727},{"id":"https://openalex.org/keywords/answer-quality-evaluation","display_name":"answer quality evaluation","score":0.3199999928474426}],"concepts":[{"id":"https://openalex.org/C203357204","wikidata":"https://www.wikidata.org/wiki/Q1089605","display_name":"Chunking (psychology)","level":2,"score":0.5656999945640564},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.4975999891757965},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46399998664855957},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.4250999987125397},{"id":"https://openalex.org/C15708023","wikidata":"https://www.wikidata.org/wiki/Q80083","display_name":"Humanities","level":1,"score":0.351500004529953},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33149999380111694},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3124000132083893},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.2879999876022339},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.26460000872612},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.25780001282691956}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:repozitorij.upr.si:IzpisGradiva.php?id=23967","doi":null,"is_oa":true,"landing_page_url":"https://repozitorij.upr.si/IzpisGradiva.php?id=23967","pdf_url":"https://repozitorij.upr.si/Dokument.php?id=35569&dn=","source":{"id":"https://openalex.org/S4377196486","display_name":"Repository of the University of Primorsk (University of Primorska)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I118905719","host_organization_name":"University of Primorska","host_organization_lineage":["https://openalex.org/I118905719"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/bachelorThesis"}],"best_oa_location":{"id":"pmh:oai:repozitorij.upr.si:IzpisGradiva.php?id=23967","doi":null,"is_oa":true,"landing_page_url":"https://repozitorij.upr.si/IzpisGradiva.php?id=23967","pdf_url":"https://repozitorij.upr.si/Dokument.php?id=35569&dn=","source":{"id":"https://openalex.org/S4377196486","display_name":"Repository of the University of Primorsk (University of Primorska)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I118905719","host_organization_name":"University of Primorska","host_organization_lineage":["https://openalex.org/I118905719"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/bachelorThesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225028353.pdf","grobid_xml":"https://content.openalex.org/works/W7225028353.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W2909544278","https://openalex.org/W2936695845","https://openalex.org/W3199367723","https://openalex.org/W4304195432","https://openalex.org/W4385565351","https://openalex.org/W4386576685","https://openalex.org/W4387156782","https://openalex.org/W4389518671","https://openalex.org/W4389984066","https://openalex.org/W4404783040","https://openalex.org/W4411113004","https://openalex.org/W4417539443"],"related_works":["https://openalex.org/W7223534785","https://openalex.org/W7206970030","https://openalex.org/W7056459307","https://openalex.org/W7141358750","https://openalex.org/W7216570215","https://openalex.org/W4413442564","https://openalex.org/W7154900482","https://openalex.org/W7125098831","https://openalex.org/W7211933033","https://openalex.org/W7172133562"],"abstract_inverted_index":{"Izvle\u010dek:":[0],"Generacija,":[1],"podprta":[2],"s":[3],"pridobivanjem":[4],"informacij":[5],"(RAG),":[6],"utemeljuje":[7],"odgovore":[8],"velikih":[9],"jezikovnih":[10],"modelov":[11],"v":[12,50,88,185],"zunanjih":[13],"dokumentih,":[14],"ki":[15],"jih":[16],"je":[17,26,127,144,215,224],"treba":[18],"pred":[19],"iskanjem":[20],"razdeliti":[21],"na":[22,94,100,136,153],"kose;":[23],"na\u010din":[24],"razdeljevanja":[25,40,152],"pomembna":[27],"in":[28,34,48,59,71,84,114,122,151,172,189,194],"premalo":[29],"raziskana":[30],"odlo\u010ditev.Ta":[31],"naloga":[32],"sistemati\u010dno":[33],"statisti\u010dno":[35,145],"utemeljeno":[36],"primerja":[37],"\u0161est":[38],"strategij":[39],"dokumentov":[41],"(fiksna":[42],"velikost,":[43],"stav\u010dno,":[44],"pomensko,":[45],"hierarhi\u010dno,":[46],"propozicijsko":[47],"kontekstualno)":[49],"\u0161tirih":[51],"domenah":[52],"(znanstveni":[53],"\u010dlanki,":[54],"pravne":[55],"pogodbe,":[56],"prepisi":[57],"sestankov":[58],"spletni":[60],"\u010dlanki),":[61],"pri":[62,167,174],"treh":[63],"metodah":[64],"pridobivanja,":[65],"ponovnem":[66],"rangiranju,":[67],"dveh":[68,72],"arhitekturah":[69],"agentov":[70],"modelih":[73],"vektorskih":[74],"vlo\u017eitev":[75],"-skupaj":[76],"132":[77],"nadzorovanih":[78],"poskusov.Kakovost":[79],"merimo":[80],"z":[81,85,106],"ogrodjem":[82],"RAGAS":[83],"jezikovnim":[86],"modelom":[87],"vlogi":[89],"ocenjevalca.Ker":[90],"vse":[91],"konfiguracije":[92],"odgovarjajo":[93],"isti":[95],"nabor":[96],"vpra\u0161anj,":[97],"primerjave":[98],"opravimo":[99],"parnih":[101],"razlikah":[102],"po":[103],"posameznih":[104],"vpra\u0161anjih,":[105],"intervali":[107],"zaupanja":[108],"za":[109,116,205],"samo":[110],"razliko,":[111],"permutacijskim":[112],"testom":[113],"popravkom":[115],"ve\u010dkratno":[117],"testiranje.Rezultati":[118],"tvorijo":[119],"delno":[120],"urejenost":[121],"ne":[123,157,173],"enotne":[124],"razvrstitve.Propozicijsko":[125],"razdeljevanje":[126,143],"izrazito":[128],"slab\u0161e":[129],"od":[130,148,160,197],"vseh":[131],"drugih":[132],"strategij,":[133],"ker":[134],"raz\u010dlenitev":[135],"atomarne":[137],"trditve":[138],"generatorju":[139],"odvzame":[140],"potrebni":[141],"kontekst.Hierarhi\u010dno":[142],"zna\u010dilno":[146],"bolj\u0161e":[147],"stav\u010dnega,":[149],"pomenskega":[150],"kose":[154],"fiksne":[155],"velikosti,":[156],"pa":[158,164],"tudi":[159],"kontekstualnega,":[161],"njegova":[162],"prednost":[163],"se":[165],"poka\u017ee":[166],"tem,":[168],"kaj":[169],"cevovod":[170],"pridobi,":[171],"kakovosti":[175],"generiranih":[176],"odgovorov;":[177],"preostale":[178],"\u0161tiri":[179],"strategije":[180],"so":[181],"med":[182,219],"seboj":[183],"ekvivalentne":[184],"mejah":[186],"0,05":[187],"to\u010dke.Gosto":[188],"hibridno":[190],"iskanje":[191],"sta":[192],"ekvivalentna":[193],"obe":[195],"bolj\u0161i":[196],"redkega":[198],"iskanja":[199],"BM25;":[200],"ponovno":[201],"rangiranje":[202],"zamenja":[203],"priklic":[204],"natan\u010dnost":[206],"brez":[207],"neto":[208],"koristi;":[209],"ve\u010dkora\u010dni":[210],"agent":[211],"redko":[212],"iterira.Prakti\u010dno":[213],"priporo\u010dilo":[214],"pogojno:":[216],"izberimo":[217],"najcenej\u0161o":[218],"zamenljivimi":[220],"strategijami,":[221],"razen":[222],"kadar":[223],"merilo":[225],"prav":[226],"kakovost":[227],"pridobivanja.":[228]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-11-01T00:00:00"}
