This dissertation investigates Slavic collective numerals and their syntactic structure from descriptive and structural perspectives on the basis of the operation Agree. The headedness of Slavic collective numeral phrases will be focused on with three Slavic languages: Russian, Bosnian/ Croatian/Serbian, and Polish. To analyze the semantic and morphosyntactic properties of Slavic collective numeral phrases, I adopted two important concepts proposed by Rappaport (2002, 2006): i) Minimal Lexical Representation (MLR) and ii) pre-valued abstract Quantitative Case (QC). MLR represents the semantic and formal features of nouns, which selectively combine with collective numerals, while the idea of QC can predict the heterogeneous and homogeneous patterns of case assignment. The presence of pre-valued abstract QC triggers heterogeneous morphosyntax, while the absence of QC triggers homogeneous morphosyntax. The spell-out forms of collective numerals are the direct result of morphological syncretic rules. In regard of the headedness of Slavic collective numeral phrases, this research claims that nouns are the heads of Slavic numeral phrases on the grounds that numerals, adjectives, and other modifiers agree with nouns, which functions as the locus of morphosyntax (Zwicky 1985). The use of collective numerals is determined by the properties of nouns. In each chapter, Slavic collective numerals will be analyzed from the three points of view: i) semantics, ii) morphology, and iii) syntax. Collective numerals can emphasize the meaning of collectivity, totality, and cohesiveness as an aggregate. BCS and Polish collective numerals strictly specify a group of mixed gender, while Russian does not. BCS is characterized by three different types of collective numerals: i) collective numeral substantives (dvojica ‘two’, trojica ‘three’, četvorica ’four’, petorica ’five’, etc.), collective numerals (dvoje ‘two’, troje ‘three’, četvoro ’four’, petoro ’five’, etc.), and collective numeral adjectives (dvoji (m.)/ dvoje (f.)/ dvoja (n.) ’two’, etc.). Moreover, indeclinability of numerals is one of the characteristics of BCS numerals. Polish has secondary gender, so-called virile marking, which does not apply to collective numerals. Polish collective numerals are strictly used to express a group of mixed gender. / text

Bakgrund: Robotprocessautomation (RPA) är en automationsteknik som kan ersätta människors arbete i IT-system. RPA växer konstant och det finns förhoppningar om att den kan bidra med positiva effekter inom hälso- och sjukvård, bland annat genom att frigöra medarbetare till andra uppgifter och öka kvalitet i form av minskade fel. Det finns behov av fler konkreta exempel av RPA inom hälso- och sjukvården. Syfte: Studiens syfte var att undersöka effekter av RPA i en sjukhusmiljö, avseende: 1) kvalitet i form av minskade fel; 2) effektivitet i form av frigjord tid för medarbetare. Metod: En experimentell design användes med kvantitativ ansats, för att undersöka två fall på ett sjukhus: 1) RPA för att minska mänskliga fel i journalsystemets läkemedelslista. Antal och andel fel som korrigerades av RPA mättes. 2) RPA för att frigöra tid för medicinska sekreterare inom en arbetsprocess. Tidsåtgång att utföra den aktuella uppgiften mättes (14 mätningar fördelat på 4 sekreterare). Antal och andel fall som hanterades av RPA mättes. Ackumulerad besparad arbetstid för organisationen mättes, tillsammans med besparad heltidsekvivalent (HTE). Resultat: 1) RPA korrigerade mänskliga misstag i 68 % av fallen (n = 3179). 2) Den manuella processen tog 27 (24,5–30,5) sekunder att utföra (median och interkvartilavstånd). Under åtta månader hanterades 9956 fall av RPA, vilket innebar frigjord tid om 9,33 (8,47–10,54) arbetstimmar i månaden inom organisationen, motsvarande 5,3 (4,8–6,0) procent av en heltidstjänst. Slutsats: Studien visade på två fall som uppnådde önskade effekter, ett fall inom kvalitet och ett inom effektivitet. Fallen utgjorde även exempel på komplicerande faktorer, som att RPA endast är temporär lösning, samt komplexitet med att delad arbetsyta mellan människa och RPA. Resultatet kan inspirera andra organisationer samt bidra till en aggregerad kunskap för att skapa riktlinjer och standarder. / Background: Robotic Process Automation (RPA) is an automation technology that can replace human work in IT systems. RPA is a growing field, and there are hopes that it can contribute positive effects in healthcare, particularly by freeing up employees for other tasks and increasing quality by reducing errors. There is a need for more examples of RPA in healthcare. Objective: The aim of this study was to investigate the effects of RPA in a hospital environment, specifically regarding: 1) quality in terms of reduced errors, and 2) efficiency in terms of freed-up time for employees. Method: An experimental design with a quantitative approach was used to examine two cases in a hospital: 1) RPA to reduce human errors in the medication list of the electronic health record system. The number and percentage of errors corrected by RPA were measured. 2) RPA to free up time for medical secretaries in a specific workflow. The time required to complete the manual task was measured through 14 measurements involving 4 secretaries. The number and percentage of cases handled by RPA were measured. Accumulated saved working time for the organization was measured, along with the saved Full-Time Equivalent (FTE). Results: 1) RPA corrected human errors in 68% of cases (n=3179). 2) The manual process took 27 (24.5-30.5) seconds to complete (median and interquartile range). During eight months, RPA handled 9956 cases, resulting in a time savings of 9.33 (8.47-10.54) work hours per month within the organization, equivalent to 5.3 (4.8-6.0) percent of a full-time position. Conclusion: The study demonstrated two cases that achieved desired effects, one in terms of quality and one in terms of efficiency. The cases also presented complicating factors, such as RPA being a temporary solution and the complexity of shared workspace between humans and RPA. The results can inspire other organizations and contribute to aggregated knowledge for creating guidelines and standards.