Robotic Process Automation (RPA) is one of the most growing technologies nowadays. RPA is an automation technology developed to replicate back-office processes and operations traditionally done by humans, thanks to software robots called bots. Bots can imitate actions, they can perform tasks receiving step-by-step instructions and they can execute these steps without breaks and with great accuracy. There are various tools used to implement bots and to automate processes, one of these is UiPath. RPA inside a business process helps companies to improve customer satisfaction, enhancing customer experience and increasing sales. Recommender systems are efficient tools for filtering online information, offering to users suitable and personalized contents or information. They are useful for e-commerce businesses, helping clients to choose the perfect product among an enormous set of options available on the online shops, based on their needs. The development of recommender systems has recently gained significant attention in the Artificial Intelligence field. This thesis aims to combine Robotic Process Automation and Artificial Intelligence by developing recommender systems using UiPath. Three types of recommender systems have been developed using UiPath: collaborative filtering, content-based filtering and hybrid filtering. We have evaluated the performance of these three methods on a dataset containing the ratings on some luxury bags given by the users. The application of Artificial Intelligence for social benefits is an emerging topic. In this thesis, the recommender systems recommend both bags and also charity fundraising campaigns that are related to the brands of the recommended bags. Therefore, these recommender systems are useful both for business processes, improving sales and customers satisfaction, and also for social good.
La Robotic Process Automation (RPA) è una delle tecnologie che più sta crescendo negli ultimi anni. RPA è una tecnologia sviluppata per l’automazione, in grado di replicare processi amministrativi e operativi tradizionalmente svolti da esseri umani questo grazie ai “software robots” chiamati anche “bots”. I “bots” eseguono e ripetono azioni senza interruzioni e con precisione, grazie a programmi in grado di fornire istruzioni dettagliate su ogni passo da compiere. Ci sono diversi strumenti per implementare i “bots” e automatizzare i processi, uno di questi è UiPath. RPA all’interno di alcuni processi di business aiuta le aziende a migliorare la soddisfazione del cliente, accrescere il gradimento del servizio offerto e aumentare le vendite dei prodotti. I sistemi di raccomandazione sono strumenti efficaci per il filtraggio delle informazioni online, offrendo agli utenti contenuti e informazioni personalizzati. Sono utili alle aziende che offrono servizi di e-commerce, per aiutare i clienti a scegliere il prodotto perfetto per le proprie necessità, all'interno di una vasta scelta di opzioni che i negozi online offrono. Lo sviluppo dei sistemi di raccomandazione ha ricevuto una grande attenzione nel campo dell'Intelligenza Artificiale. L’obbiettivo di questa tesi è combinare la Robotic Process Automation e l'Intelligenza Artificiale sviluppando sistemi di raccomadazione attraverso l'utilizzo di UiPath. Sono stati sviluppati tre sistemi di raccomadazione utilizzando UiPath: collaborative filtering, content-based filtering e hybrid filtering. Abbiamo valutato le prestazioni dai tre metodi utilizzando un dataset composto da giudizi di utenti su alcune borse di lusso. L'applicazione dell'Intelligenza Artificiale per il bene comune è una tematica emergente e sempre più importante. I sistemi di raccomandazione implementati in questa tesi raccomandano sia borse che campagne di raccolta fondi per il bene sociale lanciate dagli stessi brand delle borse suggerite precedentemente. Pertanto questi sistemi di raccomandazione sono utili sia per processi di business, che possono aiutare ad incrementare le vendite e la soddisfazione dei clienti, ma anche per scopi sociali.
Combining Robotic Process Automation and Artificial Intelligence for business processes and social good
VAROTTO, ELISA
2022/2023
Abstract
Robotic Process Automation (RPA) is one of the most growing technologies nowadays. RPA is an automation technology developed to replicate back-office processes and operations traditionally done by humans, thanks to software robots called bots. Bots can imitate actions, they can perform tasks receiving step-by-step instructions and they can execute these steps without breaks and with great accuracy. There are various tools used to implement bots and to automate processes, one of these is UiPath. RPA inside a business process helps companies to improve customer satisfaction, enhancing customer experience and increasing sales. Recommender systems are efficient tools for filtering online information, offering to users suitable and personalized contents or information. They are useful for e-commerce businesses, helping clients to choose the perfect product among an enormous set of options available on the online shops, based on their needs. The development of recommender systems has recently gained significant attention in the Artificial Intelligence field. This thesis aims to combine Robotic Process Automation and Artificial Intelligence by developing recommender systems using UiPath. Three types of recommender systems have been developed using UiPath: collaborative filtering, content-based filtering and hybrid filtering. We have evaluated the performance of these three methods on a dataset containing the ratings on some luxury bags given by the users. The application of Artificial Intelligence for social benefits is an emerging topic. In this thesis, the recommender systems recommend both bags and also charity fundraising campaigns that are related to the brands of the recommended bags. Therefore, these recommender systems are useful both for business processes, improving sales and customers satisfaction, and also for social good.File | Dimensione | Formato | |
---|---|---|---|
Varotto_Elisa.pdf
accesso riservato
Dimensione
2.17 MB
Formato
Adobe PDF
|
2.17 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/53887