Data Mining Meets HCI: Data and Visual Analytics of Frequent Patterns

2016 ◽  
pp. 289-293 ◽  
Author(s):  
Carson K. Leung ◽  
Christopher L. Carmichael ◽  
Yaroslav Hayduk ◽  
Fan Jiang ◽  
Vadim V. Kononov ◽  
...  
Keyword(s):  
Data Mining ◽  
Visual Analytics ◽  
Frequent Patterns

scholarly journals MINING STARS WITH FP-GROWTH: A CASE STUDY ON BIBLIOGRAPHIC DATA

2011 ◽  
Vol 19 (supp01) ◽  
pp. 65-91
Author(s):  
ANDREIA SILVA ◽  
CLÁUDIA ANTUNES
Keyword(s):  
Data Mining ◽  
Feasible Solution ◽  
Small Dimension ◽  
Frequent Patterns ◽  
Dimensional Model ◽  
Relational Context ◽  
Star Schema ◽  
Bibliographic Data ◽  
Massive Number

Traditional data mining approaches look for patterns in a single table, while multi-relational data mining aims for identifying patterns that involve multiple tables. In recent years, the most common mining techniques have been extended to the multi-relational context, but there are few dedicated to deal with data stored following the multi-dimensional model, in particular the star schema. These schemas are composed of a central huge fact table linking a set of small dimension tables. Joining all the tables before mining may not be a feasible solution due to the usual massive number of records. This work proposes a method for mining frequent patterns on data following a star schema that does not materialize the join between the tables. As it extends the algorithm FP-Growth, it constructs an FP-Tree for each dimension and then combines them through the records in the fact table to form a super FP-Tree. This tree is then mined with FP-growth to find all frequent patterns. The paper presents a case study on bibliographic data, comparing efficiency and scalability of our algorithm against FP-Growth.

Big Data Visualisation and Visual Analytics for Music Data Mining

2018 ◽  
Author(s):  
Katrina E. Barkwell ◽  
Alfredo Cuzzocrea ◽  
Carson K. Leung ◽  
Ashley A. Ocran ◽  
Jennifer M. Sanderson ◽  
...  
Keyword(s):  
Data Mining ◽  
Big Data ◽  
Visual Analytics ◽  
Data Visualisation ◽  
Music Data Mining

scholarly journals Preprocessing Profiling Model for Visual Analytics

2020 ◽  
Author(s):  
Alessandra Maciel Paz Milani ◽  
Fernando V. Paulovich ◽  
Isabel Harb Manssour
Keyword(s):  
Data Mining ◽  
Data Analysis ◽  
Visual Analytics ◽  
Data Preprocessing ◽  
Interview Study ◽  
Raw Data ◽  
Important Stage ◽  
Analysis Process

Analyzing and managing raw data are still a challenging part of the data analysis process, mainly regarding data preprocessing. Although we can find studies proposing design implications or recommendations for visualization solutions in the data analysis scope, they do not focus on challenges during the preprocessing phase. Likewise, the current Visual Analytics processes do not consider preprocessing an equally important stage in their process. Thus, with this study, we aim to contribute to the discussion of how we can use and combine methods of visualization and data mining to assist data analysts during the preprocessing activities. To achieve that, we introduce the Preprocessing Profiling Model for Visual Analytics, which contemplates a set of features to inspire the implementation of new solutions. In turn, these features were designed considering a list of insights we obtained during an interview study with thirteen data analysts. Our contributions can be summarized as offering resources to promote a shift to a visual preprocessing.

scholarly journals An Ontology-based Visual Analytics for Apple Variety Testing

2021 ◽  
Author(s):  
Ekaterina Chuprikova ◽  
Abraham Mejia Aguilar ◽  
Roberto Monsorno
Keyword(s):  
Data Mining ◽  
Data Analysis ◽  
Data Integration ◽  
Visual Analytics ◽  
Agricultural Sector ◽  
Environmental Data ◽  
Data Sources ◽  
Apple Variety ◽  
Testing Program ◽  
Variety Testing

<p>Increasing agricultural production challenges, such as climate change, environmental concerns, energy demands, and growing expectations from consumers triggered the necessity for innovation using data-driven approaches such as visual analytics. Although the visual analytics concept was introduced more than a decade ago, the latest developments in the data mining capacities made it possible to fully exploit the potential of this approach and gain insights into high complexity datasets (multi-source, multi-scale, and different stages). The current study focuses on developing prototypical visual analytics for an apple variety testing program in South Tyrol, Italy. Thus, the work aims (1) to establish a visual analytics interface enabled to integrate and harmonize information about apple variety testing and its interaction with climate by designing a semantic model; and (2) to create a single visual analytics user interface that can turn the data into knowledge for domain experts. </p><p>This study extends the visual analytics approach with a structural way of data organization (ontologies), data mining, and visualization techniques to retrieve knowledge from an extensive collection of apple variety testing program and environmental data. The prototype stands on three main components: ontology, data analysis, and data visualization. Ontologies provide a representation of expert knowledge and create standard concepts for data integration, opening the possibility to share the knowledge using a unified terminology and allowing for inference. Building upon relevant semantic models (e.g., agri-food experiment ontology, plant trait ontology, GeoSPARQL), we propose to extend them based on the apple variety testing and climate data. Data integration and harmonization through developing an ontology-based model provides a framework for integrating relevant concepts and relationships between them, data sources from different repositories, and defining a precise specification for the knowledge retrieval. Besides, as the variety testing is performed on different locations, the geospatial component can enrich the analysis with spatial properties. Furthermore, the visual narratives designed within this study will give a better-integrated view of data entities' relations and the meaningful patterns and clustering based on semantic concepts.</p><p>Therefore, the proposed approach is designed to improve decision-making about variety management through an interactive visual analytics system that can answer "what" and "why" about fruit-growing activities. Thus, the prototype has the potential to go beyond the traditional ways of organizing data by creating an advanced information system enabled to manage heterogeneous data sources and to provide a framework for more collaborative scientific data analysis. This study unites various interdisciplinary aspects and, in particular: Big Data analytics in the agricultural sector and visual methods; thus, the findings will contribute to the EU priority program in digital transformation in the European agricultural sector.</p><p>This project has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 894215.</p>

Association Rule Mining in Collaborative Filtering

2017 ◽  
pp. 159-179 ◽  
Author(s):  
Carson K.-S. Leung ◽  
Fan Jiang ◽  
Edson M. Dela Cruz ◽  
Vijay Sekar Elango
Keyword(s):  
Data Mining ◽  
Collaborative Filtering ◽  
Association Rules ◽  
Data Structures ◽  
Association Rule ◽  
Association Rule Mining ◽  
Real Life ◽  
Frequent Patterns ◽  
Rule Mining ◽  
Association Rule Miner

Collaborative filtering uses data mining and analysis to develop a system that helps users make appropriate decisions in real-life applications by removing redundant information and providing valuable to information users. Data mining aims to extract from data the implicit, previously unknown and potentially useful information such as association rules that reveals relationships between frequently co-occurring patterns in antecedent and consequent parts of association rules. This chapter presents an algorithm called CF-Miner for collaborative filtering with association rule miner. The CF-Miner algorithm first constructs bitwise data structures to capture important contents in the data. It then finds frequent patterns from the bitwise structures. Based on the mined frequent patterns, the algorithm forms association rules. Finally, the algorithm ranks the mined association rules to recommend appropriate merchandise products, goods or services to users. Evaluation results show the effectiveness of CF-Miner in using association rule mining in collaborative filtering.

scholarly journals BIG DATA MINING FOR INTERESTING PATTERNS WITH MAP REDUCE TECHNIQUE

2017 ◽  
Vol 10 (13) ◽  
pp. 191
Author(s):  
Nikhil Jamdar ◽  
A Vijayalakshmi
Keyword(s):  
Data Mining ◽  
Pattern Mining ◽  
Uncertain Data ◽  
Frequent Pattern Mining ◽  
Frequent Pattern ◽  
Map Reduce ◽  
Frequent Patterns ◽  
Precise Data ◽  
Big Data Mining ◽  
Transactional Databases

There are many algorithms available in data mining to search interesting patterns from transactional databases of precise data. Frequent pattern mining is a technique to find the frequently occurred items in data mining. Most of the techniques used to find all the interesting patterns from a collection of precise data, where items occurred in each transaction are certainly known to the system. As well as in many real-time applications, users are interested in a tiny portion of large frequent patterns. So the proposed user constrained mining approach, will help to find frequent patterns in which user is interested. This approach will efficiently find user interested frequent patterns by applying user constraints on the collections of uncertain data. The user can specify their own interest in the form of constraints and uses the Map Reduce model to find uncertain frequent pattern that satisfy the user-specified constraints 

Sign in / Sign up

Export Citation Format

Share Document