{"id":"https://openalex.org/W7213477126","doi":null,"title":"Evaluating Reliability in Conversational AI Agents for E-Commerce: A Comparative Study of Agent Architectures","display_name":"Evaluating Reliability in Conversational AI Agents for E-Commerce: A Comparative Study of Agent Architectures","publication_year":2026,"publication_date":"2026-07-17","ids":{"openalex":"https://openalex.org/W7213477126"},"language":"en","primary_location":{"id":"pmh:oai:repositorio-aberto.up.pt:10216/176641","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10216/176641","pdf_url":"https://repositorio-aberto.up.pt/bitstream/10216/176641/3/796384.1.pdf","source":{"id":"https://openalex.org/S4306402433","display_name":"Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"master thesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://repositorio-aberto.up.pt/bitstream/10216/176641/3/796384.1.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5151981254","display_name":"Jos\u00e9 Henrique Nogueira Campos de Morais Pinheiro","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Jos\u00e9 Henrique Nogueira Campos de Morais Pinheiro","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5151981254"],"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/T12128","display_name":"AI in Service Interactions","score":0.7827000021934509,"subfield":{"id":"https://openalex.org/subfields/1406","display_name":"Marketing"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12128","display_name":"AI in Service Interactions","score":0.7827000021934509,"subfield":{"id":"https://openalex.org/subfields/1406","display_name":"Marketing"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.1559000015258789,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.01510000042617321,"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/conversational-ai-agents","display_name":"conversational AI agents","score":0.9449999928474426},{"id":"https://openalex.org/keywords/agent-reliability","display_name":"agent reliability","score":0.7039999961853027},{"id":"https://openalex.org/keywords/e-commerce-customer-service","display_name":"e-commerce customer service","score":0.7789999842643738},{"id":"https://openalex.org/keywords/tool-use","display_name":"tool use","score":0.48500001430511475},{"id":"https://openalex.org/keywords/agent-architecture","display_name":"agent architecture","score":0.656000018119812},{"id":"https://openalex.org/keywords/multi-agent-systems","display_name":"multi-agent systems","score":0.765999972820282},{"id":"https://openalex.org/keywords/gpt-4o-mini","display_name":"GPT-4o-mini","score":0.6100000143051147},{"id":"https://openalex.org/keywords/gpt-3-5-turbo","display_name":"GPT-3.5-turbo","score":0.8840000033378601},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.6179999709129333},{"id":"https://openalex.org/keywords/user-trust","display_name":"user trust","score":0.39399999380111694},{"id":"https://openalex.org/keywords/external-tool-integration","display_name":"external tool integration","score":0.3050000071525574}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6484000086784363},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.6198999881744385},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.427700012922287},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.33649998903274536},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C74072328","wikidata":"https://www.wikidata.org/wiki/Q1142726","display_name":"Intelligent agent","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2475000023841858},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.24269999563694},{"id":"https://openalex.org/C137703981","wikidata":"https://www.wikidata.org/wiki/Q4692093","display_name":"Agent architecture","level":3,"score":0.23980000615119934}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:repositorio-aberto.up.pt:10216/176641","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10216/176641","pdf_url":"https://repositorio-aberto.up.pt/bitstream/10216/176641/3/796384.1.pdf","source":{"id":"https://openalex.org/S4306402433","display_name":"Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"master thesis"}],"best_oa_location":{"id":"pmh:oai:repositorio-aberto.up.pt:10216/176641","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/10216/176641","pdf_url":"https://repositorio-aberto.up.pt/bitstream/10216/176641/3/796384.1.pdf","source":{"id":"https://openalex.org/S4306402433","display_name":"Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"master thesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7213477126.pdf","grobid_xml":"https://content.openalex.org/works/W7213477126.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W3202712981","https://openalex.org/W4304195432","https://openalex.org/W4385965642","https://openalex.org/W4400976709","https://openalex.org/W4404057505","https://openalex.org/W4404341031","https://openalex.org/W4406892137","https://openalex.org/W4408345657","https://openalex.org/W4410427763","https://openalex.org/W4413558279","https://openalex.org/W4413917394","https://openalex.org/W4414005275","https://openalex.org/W4415821615","https://openalex.org/W7133214447","https://openalex.org/W7133222210","https://openalex.org/W7133224126"],"related_works":["https://openalex.org/W3106536562","https://openalex.org/W7213410564","https://openalex.org/W7204184376","https://openalex.org/W7214775689","https://openalex.org/W7134858195","https://openalex.org/W4416425160","https://openalex.org/W4387846864","https://openalex.org/W7163511733","https://openalex.org/W7019182600","https://openalex.org/W4391855969"],"abstract_inverted_index":{"ResumoOs":[0],"agentes":[1,134,139,230,340,475,549],"de":[2,29,32,35,37,40,43,51,92,97,112,128,133,138,141,149,179,191,198,204,211,254,257,267,272,298,302,308,321,337,355,361,369,372,374,387,414,426,428,437,464,508,512,536,540,546,550],"Intelig\u00eancia":[3],"Artificial":[4],"(IA)":[5],"conversacional":[6],"est\u00e3o":[7],"a":[8,25,27,56,59,63,67,76,82,89,94,100,109,117,122,136,145,222,246,300,315,324,331,353,367,402,410,449,472,498,506,525,528,544],"ser":[9],"cada":[10],"vez":[11,489],"mais":[12,405,422,447,521],"utilizados":[13],"no":[14,20,157,359,412,435],"contexto":[15,148],"do":[16,119,124,291,304,312,407,500],"atendimento":[17],"ao":[18,188,289,348],"cliente":[19],"com\u00e9rcio":[21],"eletr\u00f3nico":[22],"por":[23,233,238,243,260,480],"forma":[24,130],"apoiar":[26],"descoberta":[28],"produtos,":[30],"consultas":[31,256,262,362,438],"stock,":[33],"cria\u00e7\u00e3o":[34],"carrinhos":[36],"compras,":[38],"acompanhamento":[39],"encomendas,":[41],"consulta":[42],"pol\u00edticas":[44],"e":[45,58,81,121,153,168,236,264,286,323,371,409,430,454,514,527,538,570],"opera\u00e7\u00f5es":[46],"p\u00f3s-venda.Embora":[47],"os":[48,306,468,510,548],"Grandes":[49],"Modelos":[50],"Linguagem":[52],"(LLMs)":[53],"tenham":[54],"melhorado":[55],"flexibilidade":[57],"naturalidade":[60],"destes":[61],"sistemas,":[62],"sua":[64,218,220],"fiabilidade":[65,137,338,473,526],"continua":[66],"apresentar-se":[68],"como":[69,88,177],"um":[70,183,205,234,239,379,403,445],"desafio":[71],"cr\u00edtico":[72,206,408],"quando":[73],"s\u00e3o":[74,573],"chamados":[75],"interagir":[77],"com":[78,85,143,279,382,393,424],"ferramentas":[79,146,453,481],"externas":[80],"realizar":[83],"tarefas":[84,113],"v\u00e1rias":[86,462],"etapas.Erros":[87],"sele\u00e7\u00e3o":[90],"incorreta":[91],"ferramentas,":[93,299,505],"constru\u00e7\u00e3o":[95],"incompleta":[96],"argumentos,":[98],"refer\u00eancias":[99],"produtos":[101],"inexistentes,":[102],"uma":[103,161,165,169,196,385,391],"valida\u00e7\u00e3o":[104,199,370,513],"fraca":[105],"dos":[106,339,474,566],"resultados":[107,328,469,542],"ou":[108],"interrup\u00e7\u00e3o":[110],"prematura":[111],"podem":[114,460,523],"afetar":[115],"diretamente":[116],"confian\u00e7a":[118],"utilizador":[120],"qualidade":[123,556],"servi\u00e7o.Esta":[125],"disserta\u00e7\u00e3o":[126],"investiga":[127],"que":[129,207,330,441,471],"diferentes":[131],"arquiteturas":[132,156,276],"influenciam":[135],"conversacionais":[140,476,552],"IA":[142,551],"suporte":[144],"num":[147],"e-commerce.Foram":[150],"concebidas,":[151],"implementadas":[152],"avaliadas":[154,278],"tr\u00eas":[155],"mesmo":[158],"ambiente":[159],"controlado:":[160],"arquitetura":[162,166,170,173,193,223,332,376,416],"Simple":[163,174,344],"Agent,":[164],"Planner-Critic":[167,194,377],"Context-Specific":[171,224,417],"Multi-Agent.A":[172],"Agent":[175,345],"serve":[176],"base":[178,280],"compara\u00e7\u00e3o,":[180],"na":[181,281],"qual":[182],"\u00fanico":[184],"agente":[185],"tem":[186],"acesso":[187,503],"cat\u00e1logo":[189],"completo":[190],"ferramentas.A":[192],"introduz":[195],"etapa":[197],"antes":[200,216],"da":[201,217,283,317,458,495,517,557],"execu\u00e7\u00e3o,":[202],"atrav\u00e9s":[203,253,565],"avalia":[208],"as":[209,310,568,571],"a\u00e7\u00f5es":[210,572],"ferramenta":[212],"propostas":[213],"pelo":[214,486],"planner":[215],"execu\u00e7\u00e3o.Por":[219],"vez,":[221],"Multi-Agent":[225,418],"decomp\u00f5e":[226],"o":[227,294,319,335,350,383,394,420,442,455,493,502,515],"sistema":[228],"em":[229,269,287,342],"especializados,":[231],"coordenados":[232],"encaminhador":[235],"supervisionados":[237],"Stop":[240,313],"Judge,":[241,314],"respons\u00e1vel":[242],"avaliar":[244,547],"se":[245],"tarefa":[247],"foi":[248,251],"conclu\u00edda.A":[249],"avalia\u00e7\u00e3o":[250],"realizada":[252],"89":[255],"teste,":[258],"compostas":[259],"58":[261],"originais":[263],"31":[265],"variantes":[266],"par\u00e1frase,":[268],"dois":[270,433],"modelos":[271,434],"base:":[273],"gpt-3.5-turboe":[274],"gpt-4o-mini.As":[275],"foram":[277],"corre\u00e7\u00e3o":[282,354,386,427],"resposta":[284,558],"final":[285],"m\u00e9tricas":[288],"n\u00edvel":[290],"processo,":[292],"incluindo":[293],"comportamento":[295,404],"das":[296,326,519],"chamadas":[297],"estrutura":[301,507],"invoca\u00e7\u00e3o":[303],"LLM,":[305],"ciclos":[307],"revis\u00e3o,":[309],"decis\u00f5es":[311],"lat\u00eancia":[316,533],"resposta,":[318],"consumo":[320,535],"tokens":[322,537],"robustez":[325],"par\u00e1frases.Os":[327],"mostram":[329],"afeta":[333],"significativamente":[334],"perfil":[336],"baseados":[341],"LLM.O":[343],"melhorou":[346],"substancialmente":[347],"utilizar":[349],"gpt-4o-mini,":[351,395],"aumentando":[352],"0.872":[356],"para":[357,397,444],"0.956":[358],"conjunto":[360,436],"originais,":[363],"mas":[364,389,530,560],"permaneceu":[365],"vulner\u00e1vel":[366],"falhas":[368,411],"conclus\u00e3o":[373,459,518],"tarefas.A":[375],"alcan\u00e7ou":[378,419],"forte":[380],"desempenho":[381,421],"gpt-3.5-turbo,atingindo":[384],"0.960,":[388],"sofreu":[390],"degrada\u00e7\u00e3o":[392],"caindo":[396],"0.862.Esta":[398],"diminui\u00e7\u00e3o":[399],"esteve":[400],"associada":[401],"restritivo":[406],"ciclo":[413],"revis\u00e3o.A":[415],"est\u00e1vel,":[423],"pontua\u00e7\u00f5es":[425],"0.968":[429],"0.976":[431],"nos":[432],"original,":[439],"indicando":[440],"encaminhamento":[443],"dom\u00ednio":[446],"espec\u00edfico,":[448],"exposi\u00e7\u00e3o":[450],"restrita":[451],"\u00e0s":[452,504],"julgamento":[456],"expl\u00edcito":[457],"reduzir":[461],"classes":[463],"falhas.Em":[465],"termos":[466],"gerais,":[467],"demonstram":[470],"i":[477],"ii":[478],"apoiados":[479],"n\u00e3o":[482,553],"\u00e9":[483,492],"determinada":[484],"exclusivamente":[485],"modelo":[487],"base.Em":[488],"disso,":[490],"esta":[491],"resultado":[494],"intera\u00e7\u00e3o":[496],"entre":[497],"capacidade":[499],"modelo,":[501],"orquestra\u00e7\u00e3o,":[509],"mecanismos":[511],"controlo":[516],"tarefas.Arquiteturas":[520],"estruturadas":[522],"melhorar":[524],"robustez,":[529],"tamb\u00e9m":[531,561],"introduzem":[532],"adicional,":[534],"riscos":[539],"coordena\u00e7\u00e3o.Estes":[541],"destacam":[543],"import\u00e2ncia":[545],"s\u00f3":[554],"pela":[555],"final,":[559],"pelos":[562],"processos":[563],"internos":[564],"quais":[567],"respostas":[569],"produzidas.":[574]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-17T00:00:00"}