Data Analysis

This qualitative study examines the role of clients in petty corruption by analyzing actual corrupt exchanges between ordinary citizens and low level public and private employees in post-communist Hungary. Using a grounded theory approach, interviews reveal how clients from different social strata deal with low-level agents in corrupt situations. Findings suggest two contrasting forms of low-level corruption: transactions where the client and the agent do not have a prior relationship and where external factors dominate the relationship; and cases with stronger social ties between the actors, where the client has more freedom to structure the transaction. However, a client's social background frequently determines the form of corrupt transaction and the form of resources illegally exchanged in the deal.

Kernel methods have been used for various supervised learning tasks. In this paper, we present a new clustering method based on kernel density. The method does not make any assumption on the number of clusters or on their shapes. The method is simple, robust, and behaves equally or better than other methods on problems known as difficult.

"The purpose of data visualization is to offer intuitive ways for information perception and manipulation, especially for non-expert users. The Web of Data has realized the availability of a huge amount of datasets. However, the volume and heterogeneity of available information make it difficult for humans to manually explore and analyse large datasets. In this paper, we present rdf:SynopsViz, a tool for hierarchical charting and visual exploration of Linked Open Data (LOD). Hierarchical LOD exploration is based on the creation of multiple levels of hierarchically related groups of resources based on the values of one or more properties. The adopted hierarchical model provides effective information abstraction and summarization. Also, it allows efficient -on the fly- statistic computations, using aggregations over the hierarchy levels."

Recording thousands of entries during field research poses a challenge to any field researcher. Contemporary handheld computers offer affordable solutions, which can resolve this challenge. In this paper, we test the iPad tablet computer and FileMaker Go database to conduct garbological research carried out in West Bohemia (Czech Republic). Garbological research based on the collection of data about human waste requires not only efficient tools for recording a vast number of individual garbage items on the spot but also integration of multiple analytical levels in a database. Our research was aimed at household waste to illuminate consumption patterns and mobility of humans and things in contemporary Central European settings. The iPad was used to collect textual and visual data and integrate these in a relational database. We describe our methodology and experience with this kind of technology. The iPad and FileMaker Go proved to be well suited to challenging field conditions in the landfill, data collection was efficient and reliable, the database was flexible because its basic features could be modified in the field, and one could even examine preliminary trends in the data using charts in FileMaker Go. The proposed hardware and software is less efficient for the collection of precise spatial data, preparation of accurate drawings, and for projects in remote areas without good access to an electrical grid.

"Web Observatories use innovative analytic processes to gather
insights from observed data and use the Web as a platform for
publishing interactive data visualisations. Recordable events
associated with interactivity on the Web provide an opportunity to
openly evaluate the utility of these artefacts, assessing fitness for
purpose and observing their use. The three principles presented in
this paper propose a community evaluation approach to
innovation in visual analytics and visualisation for Web
Observatories through code sharing, the capturing of semantically
enriched interaction data and by openly stating the intended goals
of all visualisation work. The potential of this approach is
exampled with a set of front-end tools suitable for adoption by the
majority of Web Observatories as a means of visualising data on
the Web as part the shared, open, and community-driven
developmental process. The paper outlines the method for
capturing user interaction data as a series of semantic events,
which can be used to identify improvements in both the structure
and functionality of visualisations. Such refinements in user
behaviour are proposed as part of a new methodology that
introduces Economics as an evaluation tool for visual analytics. "