Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Os painéis de gráficos estatísticos conhecidos como dashboards são utilizados comumente naárea de Business Intelligence (BI) para a visualização de grandes sistemas organizados de dados. A presente dissertação propõe embasar o projeto de dashboards pelas teorias de Jacques Bertin, formuladas nas obras Sémiologie Graphique e La Graphique et le Traitement Graphique de linformation. Considerando este referencial, e ainda parâmetros do design de informação e da visualização de dados, foram desenvolvidos dashboards que apresentam dados sobre a política de reserva de vagas da Universidade do Estado do Rio de Janeiro, sistematizados pelo projeto de BI dessa instituição. O objetivo foi não apenas o de atender aos requisitos convencionais de um dashboard, mas sobretudo o de apresentar outras perspectivas informativas. Nesse sentido, investigam-se as especificidades dos métodos de Bertin e sua expansão para o domínio dos sistemas interativos. / The panels of statistical charts known as dashboards are employed in the area of Business Intelligence (BI) for the visualization of large-scale organized systems of data. This Masters Thesis proposes to base the design of dashboards by theories of Jacques Bertin, formulated in the books Sémiologie graphique and La graphique et le traitement de linformation. Considering this framework, and further parameters of information design and data visualization, it were developed dashboards that present data about the policy of quotas from the State University of Rio de Janeiro, systematized by the BI project of that institution. The goal was not only to meet the requirements of a conventional dashboard, but rather to present other perspectives of information. Accordingly, we investigate the specific methods of Bertin and its expansion into the area of interactive systems.

Técnicas de projeção desempenham papel importante na análise e exploração de dados multidimensionais, já que permitem visualizar informações muitas vezes ocultas na alta dimensão. Esta tese explora o potencial destas técnicas para resolver problemas relacionados à: 1) identificação de agrupamentos e 2) busca por similaridade em dados multidimensionais. Para identificação de agrupamentos foi desenvolvida uma técnica de projeção local e interativa que, além de projetar dados com ótima preservação de distâncias, permite que o usuário modifique o layout da projeção, agrupando um número reduzido de amostras representativas no espaço visual, de acordo com suas características. Os mapeamentos produzidos tendem a seguir o layout das amostras organizadas pelo usuário, facilitando a organização dos dados e identificação de agrupamentos. Contudo, nem sempre é possível selecionar ou agrupar amostras com base em suas características visuais de forma confiável, principalmente quando os dados não são rotulados. Para estas situações, um novo método para identificação de agrupamentos baseado em projeção foi proposto, o qual opera no espaço visual, garantindo que os grupos obtidos não fiquem fragmentados durante a visualização. Além disso, é orientado por um mecanismo de amostragem determinístico, apto a identificar instâncias que representam bem o conjunto de dados como um todo e capaz de operar mesmo em conjuntos de dados desbalanceados. Para o segundo problema: busca por similaridade em dados multidimensionais, uma família de métricas baseada em classes foi construída para projetar os dados, com o objetivo de minimizar a dissimilaridade entre pares de objetos pertencentes à mesma classe e, ao mesmo tempo, maximizá-la para objetos pertencentes a classes distintas. As métricas classes-específicas são avaliadas no contexto de recuperação de imagens com base em conteúdo. Com o intuito de aumentar a precisão da família de métricas classes-específicas, outra técnica foi desenvolvida, a qual emprega a teoria dos conjuntos fuzzy para estimar um valor de incerteza que é transferido para a métrica, aumentando sua precisão. Os resultados confirmam a efetividade das técnicas desenvolvidas, as quais representam significativa contribuição na tarefa de identificação de grupos e busca por similaridade em dados multidimensionais. / Projection techniques play an important role in multidimensional data analysis and exploration, since they allow to visualize information frequently hidden in high-dimensional spaces. This thesis explores the potential of those techniques to solve problems related to: 1) clustering and 2) similarity search in multidimensional data. For clustering data, a local and interactive projection technique capable of projecting data with effective preservation of distances was developed. This one allows the user to manipulate a reduced number of representative samples in the visual space so as to better organize them. The final mappings tend to follow the layout of the samples organized by the user, therefore, the user can interactively steer the projection. This makes it easy to organize and group large data sets. However, it is not always possible to select or group samples visually, in a reliable manner, mainly when handling unlabeled data. For these cases, a new clustering method based on multidimensional projection was proposed, which operates in the visual space, ensuring that clusters are not fragmented during the visualization. Moreover, it is driven by a deterministic sampling mechanism, able to identify instances that are good representatives for the whole data set. The proposed method is versatile and robust when dealing with unbalanced data sets. For the second problem: similarity search in multidimensional data, we build a family of class-specific metrics to project data. Such metrics were tailored to minimize the dissimilarity measure among objects from the same class and, simultaneously to maximize the dissimilarity among objects in distinct classes. The class-specific metrics are assessed in the context of content-based image retrieval. With the aim of increasing the precision of the class-specific metrics, another technique was developed. This one, uses the fuzzy set theory to estimate a degree of uncertainty, which is embedded in the metric, increasing its precision. The results confirm the effectiveness of the developed techniques, which represent significant contributions for clustering and similarity search in multidimensional data.

Made available in DSpace on 2016-04-29T14:23:35Z (GMT). No. of bitstreams: 1 Patricia Kunst Canetti.pdf: 6615349 bytes, checksum: 2e3fdc72fd907b8fdac9ace8c05bdb03 (MD5) Previous issue date: 2015-04-07 / This work is a survey of Canal Contemporâneo's fourteen years of memory - www.canalcontemporaneo.art.br - and analyzes this memory and its adjacent concepts to point out the prospects of this experiment / research , which reached a surprising longevity in Brazilian cultural Internet. The rescue of its history and collective memory was done in three chapters which thread runs through the editorial sections, platforms and actions of Canal Contemporâneo. In the first chapter we discuss its origin, the first stimuli, concepts and developments. Since then gathered actions that operate in the field of art, politics and communication, pointing to a perspective of narrative and rereading of contemporary art, with a work on Social Netwok Analysis and Data Visualization. The theoretical basis of this research that only begins is based on the following fields and authors: Data Visualization (Fernanda Viégas, Lev Manovich e Manuel Lima); Taxonomy (Marcia Lei Zeng e Jian Qin); Social Netwok Analysis (Katherine Faust e Stanley Wasserman) and models of Random Graphs (Paul Erdős e Alfréd Rényi), Small-World (Duncan J. Watts e Steven Strogatz), Preferential Attachment (Albert-László Barabási e Réka Albert); History and Sociology of Art (Aby Warburg, Alfred Gell e Bruno Latour). We hope that the new shared experience through this work can contribute to a broader view of collection, archiving and cultural heritage, for public policy of culture in Brazil / Este trabalho faz um levantamento da memória de quatorze anos de existência do Canal Contemporâneo www.canalcontemporaneo.art.br e analisa esta memória e seus conceitos adjacentes para apontar as perspectivas deste experimento/pesquisa, que atingiu uma longevidade surpreendente na Internet cultural brasileira. O resgate de sua história e memória coletiva foi feito em três capítulos cujo fio condutor perpassa as seções editoriais, as plataformas e as ações do Canal Contemporâneo. No primeiro capítulo abordamos a sua origem, os primeiros estímulos, conceitos e desdobramentos. Desde então reuniu ações que operam no campo da arte, da política e da comunicação, que apontam para uma perspectiva de narrativa e releitura da arte contemporânea, com um trabalho de Análise de Redes Sociais e Visualização de Dados. O embasamento teórico desta pesquisa que apenas se inicia se firma nos seguintes campos e autores: Visualizações de Dados (Fernanda Viégas, Lev Manovich e Manuel Lima); Taxonomia (Marcia Lei Zeng e Jian Qin); Análise de Redes Sociais (Katherine Faust e Stanley Wasserman) e dos modelos de Grafos Aleatórios (Paul Erdős e Alfréd Rényi), Small-World (Duncan J. Watts e Steven Strogatz), Preferential Attachment (Albert-László Barabási e Réka Albert); História e Sociologia da Arte (Aby Warburg, Alfred Gell e Bruno Latour). Esperamos que a nova experiência compartilhada através deste trabalho possa contribuir para uma visão mais ampla de acervo, arquivo e patrimônio cultural, para as políticas públicas de cultura no Brasil

Η εξέλιξη της τεχνολογίας των υπολογιστών σε συνδυασμό με την πρόοδο της τεχνολογίας των βάσεων δεδομένων έχουν συμβάλει στην ανάπτυξη νέων αποδοτικών και αυτοματοποιημένων τεχνικών για την αποτελεσματική συλλογή, αποθήκευση και διαχείριση των δεδομένων. Ως συνέπεια, ο όγκος των δεδομένων που αποθηκεύονται και είναι ευρέως διαθέσιμα ηλεκτρονικά αυξάνεται ραγδαία και η ανάγκη ανάπτυξης και χρήσης αποδοτικών μεθόδων ανάλυσης για την εξαγωγή χρήσιμης πληροφορίας καθίσταται ολοένα και πιο επιτακτική. Η εξόρυξη δεδομένων (data mining) ως ένα αναδυόμενο πεδίο διεπιστημονικών εφαρμογών συνδυάζει παραδοσιακές μεθόδους ανάλυσης δεδομένων με εξελιγμένους αλγόριθμους και διαδραματίζει σημαντικό ρόλο στην επεξεργασία μεγάλου όγκου δεδομένων. Ο όρος οπτικοποίηση δεδομένων (data visualization) αναφέρεται στη μελέτη τεχνικών οπτικής αναπαράστασης δεδομένων χρησιμοποιώντας γραφικά, κίνηση, τρισδιάστατες απεικονίσεις και άλλα πολυμεσικά εργαλεία. Στόχος των τεχνικών οπτικοποίησης είναι παρουσίαση ενός συνόλου δεδομένων με τρόπο σαφή και αποτελεσματικό που να παρέχει τη δυνατότητα εξαγωγής συμπερασμάτων και ανακάλυψης συσχετίσεων που διαφορετικά θα παρέμεναν άγνωστες. Στη διεθνή βιβλιογραφία, έχουν παρουσιαστεί αρκετές τεχνικές οπτικοποίησης δεδομένων, ενώ τα τελευταία χρόνια η επιστημονική κοινότητα έχει εστιάσει το ενδιαφέρον της και στην οπτικοποίηση των αποτελεσμάτων της εξόρυξης δεδομένων. Στα πλαίσια αυτής της μεταπτυχιακής διπλωματικής εργασίας, προτείνεται μια αποδοτική τεχνική εξόρυξης δεδομένων που βασίζεται σε γνωστές μεθόδους συσταδοποίησης, όπως ο Ιεραρχικός αλγόριθμος και o αλγόριθμος Spherical K-means και είναι κατάλληλη να εφαρμοστεί για την ανάλυση και εξαγωγή χρήσιμης γνώσης σε διαφορετικά σύνολα δεδομένων. Η προτεινόμενη τεχνική εφαρμόστηκε σε δύο διαφορετικούς τύπους δεδομένων: α) κειμενικά δεδομένα (textual data) που προέρχονται από τη βάση δεδομένων του PubMed, β) αριθμητικά δεδομένα (numerical data) από τη βάση δεδομένων της FINDbase. Επιπλέον, παρουσιάζεται μια μελέτη τεχνικών οπτικοποίησης και η ανάπτυξη σύγχρονων εφαρμογών οπτικοποίησης, τόσο για την αποτελεσματική αναπαράσταση των αρχικών δεδομένων μιας συλλογής (πριν από την επεξεργασία τους), όσο και των αποτελεσμάτων που προέκυψαν από την προτεινόμενη τεχνική συσταδοποίησης. / The evolution of computer technology along with advances in database technology have contributed to the development of new efficient and automated techniques for the effective collection, storage and management of data. As a result, the volume of stored and widely available online data is growing rapidly, and the need for effective analytical methods for extracting relevant information is becoming increasingly urgent. As an emerging field of interdisciplinary applications, data mining combines traditional data analysis methods with sophisticated algorithms and plays an important role in the processing of large volumes of data. Data visualization refers to the study of the techniques used for the visual representation of data, including graphics, animation, 3D depictions and other multimedia tools. The main goal of data visualization techniques is to present a set of data in a clear and effective way, so that the extraction of conclusions and discovery of correlations that would otherwise remain unknown, are enabled. While several data visualization techniques have been presented in the relative literature, in recent years the scientific community has been focusing on the visualization of the results obtained by the application of data mining techniques. In the present thesis, we propose an efficient data mining technique that is based on well-known clustering methods, such as the Hierarchical and Spherical K-means ones, and is suitable for the analysis and extraction of useful knowledge from different types of datasets. The proposed technique was applied into two different types of data including: a) textual data from the PubMed database, b) numerical data from the FINDbase database. Furthermore, we present a study of visualization techniques and the development of modern visualization tools for the effective representation of the original dataset (before processing) and the results obtained by the proposed clustering technique.

The aim of this compilation thesis is to understand how technology for visual storytelling can be shaped and used in relation to social science education in primary school, but also how social dimensions, technical and other matters create emerging learning conditions in such an educational setting. The visual storytelling technology introduced and used in the study is ‘the Statistics eXplorer platform, a geovisual analytics. The choice of theoretical perspectives to inform and guide the study is a socio-cultural view of human action, but also actor network theory is used to take account also of activities of technology and other matters. The study builds on three empirical materials that generate data from 16 social science teachers, and 126 students from five social science classrooms, in three Swedish primary schools. It contains field notes from the introduction of the technology; focusgroup interviews with teachers; think-aloud interviews with students and two kinds of video recordings from the classrooms (with an ordinary video camera and with software that capture activities at the computer screen, students’ activities and the audio as well). The analysis shows that the visual storytelling technology is shaped in relevant ways for social science teachers. The analysis also illustrates that the visual educational material are usable for primary school students in their social science education. They illustrate further how teachers, students, technology, information, tasks, data types, etc. together and in in close relation create highly complex learning conditions. The technology can therefore be seen as appropriate for the educational practice, but the complexity together with students’ apprehension of how to announce knowledge distribute severe problem spaces in the learning activities. The technology can therefore be assumed as a catalyst for educational change, but to achieve its potentials, reflections on didactic design and knowledge formation is requested to support the quality of students’ knowledge in relation to visual analysis. / Syftet i denna avhandling är, att förstå hur teknik för visual storytelling kan vara utformad och användas i relation till samhällsorienterande undervisning i grundskolan (årskurs 4 – 6), men också hur sociala dimensioner, tekniska och andra faktorer skapar villkor för lärande i ett sådant undervisningssammanhang. I studien introduceras datavisualiseringsteknik för visual storytelling: ‘the Statistics eXplorer platform’, ett geovisual analytics. Den teoretiska referensramen har sin grund i ett social konstruktionistiskt synsätt Ett socio-kulturellt perspektiv används för att analysera social aktivitet, men även aktörnätverks teori används för att analysera både sociala och materiella aktörer. Avhandlingen bygger på tre empiriska material som genereras med hjälp av 16 lärare i samhällsorienterande ämnen, och 126 elever tillhörande fem olika klassrum i tre olika svenska grundskolor. Materialet innehåller: fältanteckningar ifrån introduktion av tekniken, fokusgrupps-intervjuer med lärare, ‘tänka högt’-intervjuer med elever och två sorters videoinspelningar ifrån klassrum (dels med vanlig videokamera och dels med mjukvara som spelar in aktiviteter på datorskärmen och elevernas aktiviteter vid datorn, liksom ljudet). Analysen visar hur lärare, elever, teknik, information, uppgifter, data-typer, etc. tillsammans, i nära samarbete i de studerade klassrummen, skapar mycket komplexa villkor för lärande. De läraktiviteter som uppstår i klassrummen där teknik för visuell analys inkluderas, erbjuder elever support att: hantera stora datamängder, bli delaktiga i olika läraktiviteter och uppnå olika utbildningsmål, men även andra sorters elevrelaterade mål. Därför kan tekniken sägas vara relevant för denna sorts undervisning. Vidare visar analysen hur komplexiteten tillsammans med elevernas uppfattningar av hur kunskap skall visas, skapar påtagliga ‘problem spaces’ i läraktiviteterna. Lärandevillkoren kan därför förstås som en klassrumspraktik som inte fullt ut överensstämmer med den introducerade teknikens erbjudanden för visuell analys. Därför efterfrågas en förändrad syn på didaktisk design och elevers kunskapsformering, vilket blir betydelsefullt för kunskapens kvalitet i förhållande till visuell analys.

Šio darbo tyrimų sritis yra duomenų tyryba remiantis daugiamačių duomenų vizualia analize. Tai leidžia tyrėjui betarpiškai dalyvauti duomenų analizės procese, geriau pažinti sudėtingus duomenis ir priimti geriausius sprendimus. Disertacijos tikslas yra sukurti metodą tokios duomenų projekcijos radimui plokštumoje, kad tyrėjas galėtų pamatyti ir įvertinti daugiamačių taškų tarpgrupinius panašumus/skirtingumus. Šiam tikslui pasiekti yra pasiūlytas radialinių bazinių funkcijų ir daugiasluoksnio perceptrono, turinčio ,,butelio kaklelio“ neuroninio tinklo savybes, junginys. Naujas tinklas naudojamas vizualiai daugiamačių duomenų analizei, kai atidėjimui plokštumoje arba trimatėje erdvėje taškai gaunami paskutinio paslėpto neuronų sluoksnio išėjimuose, kai į tinklo įėjimą paduodami daugiamačiai duomenys. Šio tinklo ypatybė yra ta, kad gautas vaizdas plokštumoje labiau atspindi bendrą duomenų struktūrą (klasteriai, klasterių tarpusavio artumas, taškų tarpklasterinis panašumas) nei daugiamačių taškų tarpusavio išsidėstymą. / The area of research is data mining based on multidimensional data visual analysis. This allows researcher to participate in the process of data analysis directly, to understand the complex data better and to make the best decisions. The objective of the dissertation is to create a method for making a multidimensional data projection on the plane such that the researcher could see and assess the intergroup similarities and differences of multidimensional points. In order to achieve the target, a new hybrid neural network is proposed and investigated. This neural network integrates the ideas both of the radial basis function neural network and that of a multilayer perceptron, which has the properties of a ''bottleneck'' neural network. The new network is used for the visual analysis of multidimensional data in such a way that the output values of the neurons of the last hidden layer are the two-dimensional or three-dimensional projections of the multidimensional data, when the multidimensional data is given to the network. A peculiarity of the network is that the visualization results on the plane reflect the general structure of the data (clusters, proximity between clusters, intergroup similarities of points) rather than the location of multidimensional points.

The area of research is data mining based on multidimensional data visual analysis. This allows researcher to participate in the process of data analysis directly, to understand the complex data better and to make the best decisions. The objective of the dissertation is to create a method for making a multidimensional data projection on the plane such that the researcher could see and assess the intergroup similarities and differences of multidimensional points. In order to achieve the target, a new hybrid neural network is proposed and investigated. This neural network integrates the ideas both of the radial basis function neural network and that of a multilayer perceptron, which has the properties of a ''bottleneck'' neural network. The new network is used for the visual analysis of multidimensional data in such a way that the output values of the neurons of the last hidden layer are the two-dimensional or three-dimensional projections of the multidimensional data, when the multidimensional data is given to the network. A peculiarity of the network is that the visualization results on the plane reflect the general structure of the data (clusters, proximity between clusters, intergroup similarities of points) rather than the location of multidimensional points. / Šio darbo tyrimų sritis yra duomenų tyryba remiantis daugiamačių duomenų vizualia analize. Tai leidžia tyrėjui betarpiškai dalyvauti duomenų analizės procese, geriau pažinti sudėtingus duomenis ir priimti geriausius sprendimus. Disertacijos tikslas yra sukurti metodą tokios duomenų projekcijos radimui plokštumoje, kad tyrėjas galėtų pamatyti ir įvertinti daugiamačių taškų tarpgrupinius panašumus/skirtingumus. Šiam tikslui pasiekti yra pasiūlytas radialinių bazinių funkcijų ir daugiasluoksnio perceptrono, turinčio ,,butelio kaklelio“ neuroninio tinklo savybes, junginys. Naujas tinklas naudojamas vizualiai daugiamačių duomenų analizei, kai atidėjimui plokštumoje arba trimatėje erdvėje taškai gaunami paskutinio paslėpto neuronų sluoksnio išėjimuose, kai į tinklo įėjimą paduodami daugiamačiai duomenys. Šio tinklo ypatybė yra ta, kad gautas vaizdas plokštumoje labiau atspindi bendrą duomenų struktūrą (klasteriai, klasterių tarpusavio artumas, taškų tarpklasterinis panašumas) nei daugiamačių taškų tarpusavio išsidėstymą.

[pt] Esta dissertação aborda o tema do discurso persuasivo nos infográficos, um dos produtos do Design da Informação. Os infográficos são amplamente usados como ferramenta de comunicação pela mídia, com o intuito de transmitir informações de modo sintético, rápido e atraente por meio de representações visuais diagramáticas. Examina e analisa os atributos visuais dos gráficos e infográficos sob o ponto de vista da retórica visual. Investiga o uso do ferramental disponível para a elaboração de visualizações de dados persuasivas, que comuniquem de forma eloquente e eficiente o discurso desejado, obtendo a adesão do leitor. Os aportes teóricos da pesquisa se encontram na proposta de uma Retórica do Design, de Almeida Junior, fundamentada na Nova Retórica, de Chaïm Perelman e Lucie Olbrechts-Tyteca, nas investigações sobre a Percepção Visual voltada à visualização de dados, nas pesquisas de Colin Ware e Stephen Few, e na Semiótica de Charles Sanders Peirce, como elemento transdisciplinar, perpassando pelos diferentes tópicos como forma de integrá-los. Foram tomados, como casos exemplares, infográficos da seção Jornais da Sexta Mostra Nacional de Infografia de 2012, o Infolide. Os infográficos analisados mostram a intensa presença de recursos e ferramentas de persuasão na infografia impressa. Como resultado, chegou-se à conclusão de que é possível o designer intensificar o poder persuasivo dos seus infográficos, aprofundando-se acerca dos sistemas cognitivos da linguagem que regem a compreensão do leitor, ou seja, do seu auditório. / [en] This dissertation addresses the topic of persuasive speech in infographics, one of the products of the Information Design. The infographics are widely used as a communication tool by the media, in order to transmit information in a synthetic, fast and attractive way using visual diagrammatic representations. Examines and analyzes the visual attributes of the data graphs and infographics from the point of view of visual rhetoric. Investigates the use of the tools available for developing compelling data visualizations that communicate eloquently and efficiently the desired speech, with the adherence of the reader. The theoretical references of the research are the proposal of a Rhetoric of Design, by Almeida Junior, based on the New Rhetoric of Chaïm Perelman and Lucie Olbrechts- Tyteca, the investigations on Visual Perception oriented to data visualization, in surveys of Colin Ware and Stephen Few, and the Semiotics of Charles Sanders Peirce, as a transdisciplinary element, passing through the different topics, integrating them. Were taken, as special cases, infographics from the Newspapers section of the 6th National Exhibition Infographics 2012, Infolide. The analyzed infographics show the intense presence of persuasive tools in printed infographics. As a result, the conclusion that the designer can enhance the persuasive power of their infographics deepening his knowledge about cognitive and language systems that govern the reader s understanding, ie, of his audience.

The increased interest for self-tracking through the use of technology has given birth to the Quantified Self movement. The movement empowers users to gain self-knowledge from their own data. The overall idea is fairly recent and as such it provides a vast space for exploration and research. This project contributes to the Quantified self movement by proposing a concept for visualization of personal data using an avatar. The overall work finds inspiration in Chernoff faces visualization and it uses parts of the presentation method within the project design.   This thesis presents a visualization approach for Quantified Self data using avatars. It tests the proposed concept through a user study with two iterations. The manuscript holds a detailed overview of the designing process, questionnaire for the data mapping, implementation of the avatars, two user studies and the analysis of the results. The avatars are evaluated using Spotify data. The implementation offers a visualization library that can be reused outside of the scope of this thesis. The project managed to deliver an avatar that presents personal data through the use of facial expressions. The results show that the users can understand the proposed mapping of data. Some of the users were not able to gain meaningful insights from the overall use of the avatar, but the study gives directions for further improvements of the concept. / Visualizing quantified self data using avatars

Made available in DSpace on 2016-06-02T19:06:12Z (GMT). No. of bitstreams: 1 5971.pdf: 5628502 bytes, checksum: 38bfe45912e4f91f4ad8c7fb5fb815db (MD5) Previous issue date: 2012-08-03 / Universidade Federal de Minas Gerais / Technological development brought improvements in the technology of climate sensors and Earth's surface image acquisition, gathering increasing amounts of data. Generally, when these data are submitted to mining algorithms, the output is the production of hundreds or even thousands of textual patterns, making the task of data analysis by the domain expert even harder. Hence, it is crucial, to support experts, the development of a tool that helps to identify and display patterns of interest. In this context, this research project at Master Science level aims to develop a technique for mining association rules in time series allowing agrometeorological data analysis over time. / O avanço tecnológico tem propiciado melhorias nos diversos sensores utilizados para medições dos dados climáticos e de imageamento da superfície terrestre, coletando quantidades cada vez maiores de dados. Quando esses dados são submetidos aos algoritmos de mineração para serem explorados ocorre, em geral, a produção de centenas ou ate mesmo milhares de padrões textuais, dificultando ainda mais a tarefa de analise dos dados pelo especialista de domínio. Assim, e crucial, para apoiar os especialistas, o desenvolvimento de um ferramental que auxilia na identificação e visualização dos padrões de interesse. Neste contexto, este projeto de pesquisa em nível de mestrado visa desenvolver uma técnica de mineração de regras de associação em series temporais permitindo a analise de dados agrometeorológicos ao longo do tempo.