Sarajane Marques Peres

• Ph.D
• Professor (Associate) at University of São Paulo

The predominant focus of research in conformance checking does not account for the likelihood of behaviors allowed by the process model. If multiple activities are enabled based on the current state of the process model, they are assumed to have equal probabilities of occurring. However, some could be more probable than others. Existing approaches...

Machine learning (ML) interpretability is vital for advancing Predictive Process Monitoring (PPM) and making ML more actionable. The VisInter4PPM framework was designed to bridge the interpretability gap in PPM by visualizing insights on process predictions. This study evaluates VisInter4PPM’s effectiveness in offering interpretable predictions for...

Although large language models (LLMs) demonstrate strong text generation capabilities, they struggle in scenarios requiring access to structured knowledge bases or specific documents, limiting their effectiveness in knowledge-intensive tasks. To address this limitation, retrieval-augmented generation (RAG) models have been developed, enabling gener...

Predictive process monitoring (PPM) faces fairness issues due to biases in historical data, causing discriminatory practices. Balancing fairness and performance in PPM is crucial but underexplored, possibly requiring the adaptation of ML fairness techniques to process mining. This study assesses Reweigh-ing, Adversarial Debiasing, and Equalized Odd...

Large language models have changed the way various applications are developed. Interactions with large language models have reached a new level of complexity and now act as real problem solvers. However, despite their apparent competence, it is still necessary to accredit them with respect to the tasks they are assigned. In this paper, we discuss a...

Background: Depression in older adults is a prevalent issue that can lead to severe consequences including a decline in overall health and even suicide. Early detection and management of depression are crucial for preventing such outcomes. The integration of technology solutions in healthcare represents a promising approach to support prevention ,...

Although large language models (LLMs) demonstrate strong text generation capabilities, they struggle in scenarios requiring access to structured knowledge bases or specific documents, limiting their effectiveness in knowledge-intensive tasks. To address this limitation, retrieval-augmented generation (RAG) models have been developed, enabling gener...

Resumo: Pessoas institucionalizadas, especialmente as que apresentam processos demenciais, geralmente têm pouco envolvimento em atividades sociais e lazer. A robótica socialmente assistiva pode atuar na minimização do isolamento social nessa população. Objetivos: Relatar uma experiência inovadora de utilização de um robô socialmente assistivo para...

Knowledge graphs are employed in several tasks, such as question answering and recommendation systems, due to their ability to represent relationships between concepts. Automatically constructing such a graphs, however, remains an unresolved challenge within knowledge representation. To tackle this challenge, we propose CtxKG, a method specifically...

Pirá is a reading comprehension dataset focused on the ocean, the Brazilian coast, and climate change, built from a collection of scientific abstracts and reports on these topics. This dataset represents a versatile language resource, particularly useful for testing the ability of current machine learning models to acquire expert scientific knowled...

Coping with concept drifts in process mining remains highly pertinent, given the inherently dynamic essence of real-world business processes. Daily operations see processes undergoing continuous changes due to changing market demands, technological advances, or organizational restructuring. Conceptual drifts may emerge unexpectedly at any time, imp...

Este texto pretende apontar para as oportunidades da mineração de processos como ferramenta para apoiar a administração pública no cumprimento de seus valores e princípios, bem como atender às exigências atuais de digitalização, de novos modelos de formulação de políticas públicas baseadas em dados, de desempenho, compliance e transparência de seus...

We consider a generalization of the task allocation problem. A finite number of human resources are dynamically available to try to accomplish tasks. For each assigned task, the resource can fail or complete it correctly. Each task must be completed a number of times, and each resource is available for an independent number of tasks. Resources, tas...

Advances in sign language processing have not adequately kept pace with the tremendous progress that has been made in oral language processing. This fact serves as motivation for conducting research on the potential utilization of deep learning models within the domain of sign language processing. In this paper, we present a method that utilizes de...

Information yielded by unsupervised learning is often hard to interpret due to the lack of defined labels. To overcome this, we propose and illustrate a strategy for interpreting and visualizing the results of coclustering algorithms based on trifactorization. Our method consists of three steps: (1) vector space visualization; (2) cluster character...

Conversational agents can now operate with language models, rules, ontologies and varied other sources to provide smooth dialogue. However, the coordination of multiple sources in conversational agents is a challenge. We present a mechanism to effectively orchestrate multiple sources in a conversational agent, by relying on a client-server approach...

A linguagem neutra está no centro de discussões sobre inclusão e combate a vieses de gênero. Pautada na neutralização de gênero, ela é caracterizada pela adição de novos elementos de gênero neutro em uma língua, ou pela priorização da escrita em sintaxe neutra. Ambas as formas são processáveis automaticamente e podem ser tratadas no escopo do proce...

Resumo: A transparência pública possibilita o exercício da democracia tanto pela orientação a melhores prá-ticas de gestão por parte das instituições públicas como pela promoção da participação ativa dos cidadãos na gestão pública. A transparência é uma conduta essencial em um contexto democrático e sua prática deve ser continuamente incentivada e...

A transparência pública possibilita o exercício da democracia tanto pela orientação a melhores práticas de gestão por parte das instituições públicas como pela promoção da participação ativa dos cidadãos na gestão pública. A transparência é uma conduta essencial em um contexto democrático e sua prática deve ser continuamente incentivada e aperfeiço...

Contemporary process mining techniques commonly assume business processes are in a steady state. However , business processes are prone to change and evolution in response to various factors, which can happen at any time, in a planned or unplanned way. This phenomenon of business process evolution and change is known as concept drift, and identifyi...

The majority of the state-of-the-art predictive process monitoring approaches are based on machine learning techniques. However, many machine learning techniques do not inherently provide explanations to business process analysts to interpret the results of the predictions provided about the outcome of a process case and to understand the rationale...

A automação da descoberta de um modelo de processo de negócio e da verificação de conformidade são os tipos de mineração de processos mais populares. Presentes como funcionalidades na maioria das ferramentas existentes para a área, eles são tarefas que revelam informações importantes sobre um processo e apoiam o trabalho eficiente e eficaz de gesto...

A mineração de processos, assim como outras disciplinas inspiradas na mineração de dados, usa algoritmos de aprendizado de máquina para incrementar suas funcionalidades. As soluções em mineração de processos aproveitam o potencial do aprendizado de máquina para reduzir dificuldades inerentes a tarefas complexas. Como resultado, a proposição de solu...

Object-centric event log is a format for properly organizing information from different views of a business process into an event log. The novelty in such a format is the association of events with objects, which allows different notions of cases to be analyzed. The addition of new features has brought an increase in complexity. Clustering analysis...

Smart toys pose potential privacy risks, which may lead to aversion of parents who may choose not to buy them for their children. Magazine and newspaper articles have recently discussed these risks. However, in developing economies, these toys are not yet widely marketed, and hence the result may be different, perhaps due to potential consumer igno...

In recent years, experts in theory of gesture have been showing some interest in automating the discovery of gesture information. Such an automation can help them in reducing the inherent subjectivity of gesture studies. Usually, to produce information for linguistic and psycholinguistic studies, the researchers analyze a video of people speaking a...

In many organizational contexts, the existence of a normative process model makes it possible to verify if the actual execution of activities of a business process conforms to that model. Non-conforming behavior can be detected by aligning the actions recorded in the event log with a related normative process model. The alignment approach uses a co...

Question answering (QA) systems are usually structured as strict conditional generators, which return an answer for every input question. Sometimes, however, the policy of always responding to questions may prove itself harmful, given the possibility of giving inaccurate answers, particularly for ambiguous or sensitive questions; instead, it may be...

Reports of errors committed in public contexts by facial recognition systems based on machine learning techniques have multiplied. Still, these systems have been increasingly used by the Brazilian public administration. Consequently, the following key problem is established: how can errors committed by facial recognition systems be prevented or mit...

Resource allocation to execute business processes is increasingly crucial for organizations. As the cost of executing process tasks relies on several dynamic factors, optimizing resource allocation can be addressed as a sequential decision process. Process mining can aid this optimization with the use of data from the event log, which records histo...

Object-centric event log is a format for properly organizing information from different views of a business process into an event log. The novelty in such a format is the association of events with objects, which allows different notions of cases to be analyzed. The addition of new features has brought an increase in complexity. Clustering analysis...

Pirá is a recently developed reading comprehension dataset focused on the ocean, the Brazilian coast, and climate change. No detailed set of baselines has been built with this dataset yet, something that certainly hinders its use by researchers. In this paper, we define five benchmarks over the Pirá dataset, covering machine reading comprehension,...

We describe the first steps in the development of an artificial agent focused on the Brazilian maritime territory, a large region within the South Atlantic also known as the Blue Amazon. The "BLue Amazon Brain" (BLAB) integrates a number of services aimed at disseminating information about this region and its importance, functioning as a tool for e...

Knowledge graphs are employed in several tasks, such as question answering and recommendation systems, due to their ability to represent relationships. Manually constructing knowledge graphs, however, may represent a complicated and costly activity. Thus, one of the key challenges in the knowledge representation field has been that of automatically...

When it comes to process mining, the more singularities there are in a business process, the more human-in-the-loop strategies are needed. In order to apply discovery, compliance, profile identification, prediction or recommendation algorithms, domain experts are often involved in various tasks, such as pre-processing event logs, parameterizing alg...

Several studies have shown valuable results in remaining time prediction. However, the analysis of non-fitting cases and their impact on the prediction accuracy have been carried out superficially. Non-fitting cases are those for which there is no full match for a new case presented to the predictor. We analyzed the impact of non-fitting cases on a...

Smart toys are becoming increasingly present in children's lives, reinforcing the relevance of this market niche. Advances in user interfaces and artificial intelligence have been incorporated into smart toys to provide greater autonomy and inductive reasoning skills through machine learning. However, machine learning embedded in smart toys not onl...

Public transparency enables the exercise of democracy by the active participation of citizens in the public management. Even though promoting transparency is an essential conduct in a democratic context, its practice is still incipient. In this context, process mining emerges as an agent to promote public transparency as the data related to public...

Current smart toy parental control tools offered by toy companies do not adequately support parents in protecting their children. Moreover, there is no reference solution in the literature to be used by toymakers. Most studies are limited to mentioning the importance and purpose of these tools or some specific requirements. This article proposes a...

Current research in natural language processing is highly dependent on carefully produced corpora. Most existing resources focus on English; some resources focus on languages such as Chinese and French; few resources deal with more than one language. This paper presents the Pirá dataset, a large set of questions and answers about the ocean and the...

Time-based prediction problems are often modeled using machine learning. In business process monitoring, we associate time-based prediction tasks with predictive process monitoring goals. Solutions for prediction are based on typical pieces of information recorded in an event log related to business processes, such as timestamps and execution of ac...

Flipped classroom is an active learning method that encourages students to access study material prior to class time. Ensuring the flipping process took place, understanding how it occurred, and verifying whether it produced positive results has been a challenge for lecturers. In this article, we analyze a flipped classroom scenario through process...

Atualmente, pesquisadores em mineração de processos têm se preocupado em ir além da análise de processos isolados, uma vez que a complexidade dos processos organizacionais exige análise de múltiplos processos inter-relacionados. Mineração de processos centrada em objetos traz uma proposta conceitual para lidar com esse tipo de situação. Contudo, tr...

Process mining explores event logs to offer valuable insights to business process managers. Some types of business processes are hard to mine, including unstructured and knowledge-intensive processes. Then, trace clustering is usually applied to event logs aiming to break it into sublogs, making it more amenable to the typical process mining task....

The use of smart toys by children raises new concerns for parents and researcher. Children are more likely to share sensitive data and are unaware or rarely care about online risks. Parents play a relevant role in protecting the children, and parental control tools are needed to take control and adequately manage their child's data according to the...

In recent years, machine learning has been used for data processing and analysis, providing insights to businesses and policymakers. Deep learning technology is promising to further revolutionize this processing leading to better and more accurate results. Current trends in information and communication technology are accelerating widespread use of...

Electroencephalogram (EEG) is a non-invasive tool used to monitor the electrical activities of the brain. EEG signal analysis has several applications in the medical field. It is widely used for clinical diagnostics and for advances in the Brain-Computer Interface (BCI) area. In recent years, several studies about the automatic execution of this an...

Purpose This review scopes evidence on the use of social robots for older adults with depressive symptoms, in the scenario of smart cities, analyzing the age-related depression specificities, investigated contexts and intervention protocols' features. Design/methodology/approach Studies retrieved from two major databases were selected against incl...

We present in this paper a scoping review conducted in the interactive clustering area. Interactive clustering has been applied to leverage the strengths of both unsupervised and supervised learning. In interactive clustering, supervised learning is represented by inserting the knowledge of human experts in an originally unsupervised data analysis...

Aim: A smart toy robot has its intellect with circuits on board. It has a built-in microprocessor, sensors of one or more types, a mechanical system including moving parts, and some firmware to control and tie the parts together. The embedded sensors and devices help to create their functionality. These devices include wireless communication for da...

Business processes allow anomalies to occur during execution. Anomaly detection aims to discover behaviors that are not typical or expected in the business process. In fact, early detection helps prevent intrusion and other risks in companies. There are several approaches that address this problem in process mining. This paper discusses anomaly det...

This article describes the main components of the research theoretical framework on process mining for legal compliance. Based on the research gap amid this theoretical framework, the research objective is defined and a research method is proposed to achieve this objective.

Most process mining techniques assume stationary processes and are not well equipped to deal with concept drift. Online detection, localization and characterization of concept drift in business processes can support process mining techniques and analysts to improve organizations flexibility and adaptability. In this research, we propose a method to...

Smart toys raises new concerns for parents and researchers. Children are more likely to share sensitive data and are unaware or rarely care about online risks. Parents play a relevant role in protecting the children, and parental control tools are necessary to take control and properly manage their child's data, according to their preferences. Howe...

Coclustering algorithms are an alternative to classic one-sided clustering algorithms. Because of its ability to simultaneously cluster rows and columns of a dyadic data matrix, coclustering offers a higher value-added information: it offers column clusters besides row clusters, and the relationship between them in terms of coclusters. Different st...

A baixa representatividade feminina na área de tecnologia motiva, globalmente, iniciativas para diminuir a disparidade de gênero na área. Este artigo apresenta o projeto GRACE, cujo objetivo é fomentar a participação feminina na área da Computação, e discute sua iniciativa seminal: levar informação sobre computação para meninas de oitavo e nono ano...

Nowadays, natural language processing techniques enable the development of applications that promote communication between humans and between humans and machines. Although the technology related to automated oral communication is mature and affordable, there are currently no appropriate solutions for visualspatial languages. In the scarce efforts t...

Business process descriptions are useful documents that are becoming increasingly important for identifying and documenting business processes. They are particularly beneficial during discovery when information about the process is gathered in interviews or by observation. Such business process descriptions are written as natural language text, whi...

Process mining aims to automatically discover, analyze and improve business processes. Trace clustering is a task commonly used to reduce the inherent complexity of processes by identifying patterns. This research focuses on the application of experts knowledge in process mining through interactive clustering, referred to herein as interactive trac...

A descoberta e caracterização de perfis de comportamento de condução pode ser útil para apoiar a otimização de processos para seguradores ou gestoras de frotas. A evolução das técnicas da computação ubíqua e de análise de dados tornou tais tarefas possíveis. Neste artigo, nós apresentamos um estudo na análise de comportamentos de condução por meio...

Biclustering and coclustering are data mining tasks capable of extracting relevant information from data by applying similarity criteria simultaneously to rows and columns of data matrices. Algorithms used to accomplish these tasks simultaneously cluster objects and attributes, enabling the discovery of biclusters or coclusters. Although similar, t...

Surprise is a property of recommender systems that has been receiving increasing attention owing to its links to serendipity. Most of the metrics for surprise poorly agree with definitions employed in research areas that conceptualise surprise as a human factor, and because of this, their use in the task of evaluating recommendations may not produc...

Approaches have been proposed in process mining to predict the completion time of process instances. However, the accuracy levels of the prediction models depend on how useful the log attributes used to build such models are. A canonical subset of attributes can also offer a better understanding of the underlying process. We describe the applicatio...

Machine Learning Basic Concepts - Graphical Summary