Existing Web interoperability protocols are, arguably, overly complex as a result of each protocol being designed by a different group, providing a single service, and having its own syntax and vocabulary. Some standards, such as RSS, are popular and are designed with simplicity in mind and include easy to understand documentation, which is a key reason for its high adoption levels. However, the majority of protocols are complex, making them relatively difficult for programmers to understand and implement and thus hampering communication between heterogeneous information systems. This paper proposes a possible new direction for high-level interoperability protocols by focusing on simplicity. The High-level Interoperability ProtocolCommon Framework (HIP-CF) was designed and evaluated as a proof of concept that, if interoperability is simplified and it is made easier for programmers to understand and implement protocols, it could lead to having more interoperable systems as well as increased protocol adoption levels. Evaluation showed that this is a reasonable view and that there is a lot of room for improvement when it comes to interoperability protocols.

The Bleek and Lloyd collection contains 19th century handwritten notebooks that document the language and culture of the |Xam-speaking people who lived in Southern Africa. Access to this rich data could be enhanced by transcriptions of the text; however, the complex diacritics used in the notebooks complicate the process of transcription. Machine learning techniques could be used to perform this transcription, but it is not known which techniques would produce the best results. This paper thus reports on a comparison of 3 popular techniques applied to this problem: artificial neural networks (ANN); hidden Markov models (HMM); and support vector machines (SVM). It was found that an SVM-based classifier using histograms of oriented gradients as features resulted in the best word recognition accuracy of 58.4%. Furthermore, it was found that most feature extraction parameters did not have a large effect on recognition accuracy and that the SVM-based recognisers outperform both ANN-and HMM-based recognisers.

No abstract