Mohd Syafiq Zolkeply
Affix & Matching Model (ANM) for Iban Language Stemmer.
Context: As a multi-racial country, Malaysia is known for its indigenous ethnics primarily located in West Malaysia. It was calculated, there are twenty-seven ethnics alone in Sarawak, a major state in West Malaysia. The biggest community of this ethnic is Iban. The Ibanese communicate through Iban language, which is one of the native languages to preserve. Problem: Unfortunately, current Information Retrieval (IR) system does not support any query related to Iban language. When of Iban word is queried, the results obtained are unrelated. Proposed Approach: Therefore, we propose an Affix and Matching Model (ANM) to enhance current Information Retrieval (IR) system to retrieve information pertaining Iban language. ANM works by recognizing Iban word and later remove any unnecessary prefix and suffix using stemming algorithm only to capture the root word. Using the String matching algorithm (ESMA), the root word is matched with the word that has been classified respectively in the repository. Once the match is found, it gives the exact meaning and later it will affix together with the suffix and prefix that were removed earlier to give more meaningful result. The evaluation of this model is based on the accuracy of result retrieved when implementing in the information retrieval system environment.
Mohd Syafiq Zolkeply was a lecturer in the Faculty of Computer and Mathematical Sciences at Universiti Teknologi MARA (UiTM), Malaysia. He was conferred with Master of Software Engineering at Universiti Teknologi Malaysia (UTM) and BSc in Software Engineering at Universiti Kuala Lumpur (UniKL). He previously worked as an IT Executive in the Enterprise Transformation Service (ETS) Division at Malayan Banking Berhad (Maybank). Prior to that, he worked as a Software Engineer at various software houses in Malaysia for three years. Currently he is pursuing his PhD in the School of Computer Science and Informatics at Cardiff University. He is supervised by Dr Jianhua Shao and Dr Wendy Ivins. His main area of interest is in the evolution of software system particularly mining software repositories using machine learning techniques. Software repositories such as bug reports contain rich information about the current state of software’s behaviour. Leveraging such reports will provide insights into software systems, problems commonly faced during software development and predict the growth of complex software systems. By applying machine learning techniques to analyse the bug reports, would help both software practitioners and researchers to observe interesting patterns in software evolution for better growth.
Communities and the computational: Researching popular music experiences.
This paper will demonstrate how I have used approaches from the emerging fields of Digital Humanities and Cultural Analytics in my analysis of 8000 stories gathered from a community of music listeners by The Harkive Project (www.harkive.org). In doing so I will explore some of the issues and questions raised by the emerging role of data-related technologies and techniques in Arts and Humanities research.
Having employed similar types of computational/algorithmic collection and processing to those used by key players in the digital music space, the aim of my research has been to understand how data-derived knowledge creation works through practice, and has thus been an attempt to ‘peer under the hood’ of algorithmic and computational processes as they relate to popular music culture.
By reflecting upon the process of using these technologies and techniques, my paper will explore the limitations and affordances of undertaking cultural scholarship in this way. Such approaches not only invite us to think about how we might develop new research skills, but to simultaneously challenge many widely held assumptions related to, for instance, what a data point can and does represent. We may also examine the potential impacts on communities of interest when social action is reduced to data points, which in turn can be subjected to statistical and other mathematical analyses that often under-privilege anomalies and outliers.
In addition to the above, and in consideration of Sandvig and Hargattai’s recent work highlighting the importance of ‘benchwork’ (2015), my paper will be accompanied by related code scripts, sample data sets, an instructional blog post, and video, that together may enable scholars to replicate and/or build upon my work.
Craig Hamilton is an AHRC Midland3Cities-funded PhD research student at the School of Media at Birmingham City University, due to complete his thesis in September 2017. His research focus is the experience of contemporary popular music listeners, and specifically the relative and interrelated impacts of emerging technologies on the business and cultural environments of music consumption. He is exploring this through the development of The Harkive Project (www.harkive.org), which aims to gather reflections from music listeners about the detail of their practice by asking people to describe How, Where and When they listen to music across the course of a single day. Since 2013 the Harkive project has gathered around 8,000 stories through a number of online channels, ranging from short tweets to long-form essays. Together these text-based stories and images form a complex and rich dataset that describes contemporary popular music consumption practices and the role of music in the everyday lives of participants. In order to help derive insight from this corpus Craig has sought to develop an innovative approach, and largely this has been based on the use of computational analytical processes that are similar in nature to those increasingly coming to bear on the popular music experience itself. The aim of this has been to gain an understanding of the contents of the corpus that a ‘human’ reading may not be able to arrive at, develop a better understanding of the role of data and digital technologies on the popular music experience, and to explore issues and questions related to the use of computational techniques in media and cultural research.
Building Digital Learning Communities: The Victorian Illustrated Shakespeare Archive.
The Victorian era was the ‘Golden Age’ for Shakespeare illustration. Between 1839 and 1880 thousands of illustrations were produced within many different editions of Shakespeare’s Complete Works. What is so fascinating about these illustrations is that they have, historically, been widely neglected by academic scholarship. These editions, which were hugely popular in the Victorian era, are a very important part of our cultural heritage and, indeed, our construction of Shakespeare’s plays as we understand them today.
The Victorian Illustrated Shakespeare Archive is centred on the four major Victorian illustrated editions of Shakespeare’s Complete Works and makes available online over 3000 of these illustrations in an open-access database. The archive will be available online at ‘ShakepseareIllustration.com’ and will allow researchers and members of the public to explore a rich image archive and to ask new questions about this material: for example, ‘how did the Victorians portray certain characters and plays pictorially and does this portrayal differ throughout the Victorian era?’
Alongside such questions, the archive, more broadly, allows users to explore and interrogate the complex relationship that exists between the page and the stage, between word and image and between the past and the present. Underpinning the project is my strong belief that an online academic resource can be both scholarly rigorous and user-friendly. Further, the archive uses social networking to enable a community of users to discuss the images and to collaborate in exciting new and unforeseen ways.
This paper will explore the archive and the way it is being used across many different educational environments.
Michael Goodman, completed his PhD at Cardiff University in December 2016. His thesis, ‘Illustrating Shakespeare: Practice, Theory and the Digital Humanities’ explored how digital technology can be used to make sense of historical (specifically Victorian) illustrations of Shakespeare’s plays. The project saw the launch of the Victorian Illustrated Shakespeare Archive, an online open access resource that contains over 3000 illustrations taken from Victorian editions of Shakespeare’s plays. A founding member of Forms of Innovation (an AHRC collaborative project that investigated the interplay between technology and literature), Michael has also worked on the forthcoming Women in Trousers: A Visual Archive, designed the online medical humanities gallery Visualising the Condition and Experience of Seizures and is on the advisory board of the new Wellcome Trust-funded ‘Science Humanities’ initiative at Cardiff University. He was also the RA on Cardiff University’s Digital Culture’s Network.
Communication networks, windfarms and the common world to be built.
Using a common land dispute as a case study, this research examines interactions within and between two organisations who have fought over two decades for political, legal and physical control of a Welsh mountain. One of them, a multinational energy company wishes to build a windfarm while the other, a community protest group, wants the space to remain undeveloped. The dispute has been intertwined with the process of obtaining planning permission and common land exchange swaps. The respective beliefs and practices of the factions have been elucidated through both representations in the local news environment, their internal communications. Frames of understanding the dispute at work in these texts through source and argument deployment are tools used by campaign members and leaders to make sense of situations and provide a basis for action. The study takes this interplay of the semiotic and the material as the fundamental means by which differing versions of stability are sought. Content analysis is used to infer communications practice and organisational strategy. Interviews with campaigners, politicians, scientists and journalists locate the socio-historical context of the protest and it’s reporting. They also describe affective responses generated in relationship to landscape. Actor Network Theory is used to conceptualise the case as a series of interlocking networks including executive and juridical power centres, global capital, digital communications and medieval systems of land ownership. This mix finds expression in the competing factions but also exists in contested iterations of how the public constitutes itself democratically and what then it might want.
Matthew Pudner is a Journalism Studies PhD candidate at JOMEC, Cardiff University. His research concerns the connection between local journalism, civic life and protest. It examines information ecologies figured as networks in which government, corporate, rentier and community stakeholders engage in trials of strength to control land. He recently studied attempts to abrogate common land rights in rural Swansea allowing construction of a windfarm. The study traces the effectiveness and durability through the record of digital networks, showing how a quarter century of political, legal and direct struggle was affected by journalistic practice and technological materiality. It shows how the development was passed: Through strategic communication aiming to unspool postulations hitherto and popularly accepted as factual, exertion of influence over the application processes and work to mitigate public discussion of that influence. Matthew was an undergraduate in Cardiff. He holds a postgraduate certificate in teaching adult education and Master of Science degrees in Special Education and Social Science Research Methods. He was a classroom teacher at P369K in beautiful downtown Brooklyn for six years. Since then he has taught at Gower College and Swansea University. He enjoys the action cinema of Noam Chomsky and the philosophy of Jean Claude Van Damme. Every chance to split firewood is seized upon.