This paper focuses on investigating factors affecting e-revenue adoption in State Internally Revenue Service. The study utilizes a quantitative research methods. A conceptual research model to investigate factors affecting e-revenue was developed by integrating technology, organisation, and environment framework. The constructs employed in predicting e-revenue adoption include technological competence, financial cost, internal need, satisfaction with existing system, competitive pressure, taxpayer readiness, government regulation. Data were collected from 140 staff of the ICT department, collection departments, and some management staff of State Internal Revenue Service in three state of Nigeria. The data were analysed based on PLS-SEM using SmartPLS 3.0. The result shows that financial cost, level of satisfaction with existing system, internal need of the revenue agencies, government regulation, and competitive pressure are significant factors influencing the adoption of e-revenue in Nigeria.

No abstract

Parallel text is a fundamental requirement for the development of corpus based or data driven machine translation systems and other Natural Language Processing applications. The unavailability of this valuable linguistic resource has greatly hampered the development of NLP applications in English and Igala language. This study is aimed at creating EnglishIgala parallel text. The result of the study in addition to providing linguistic resource will enhance language learning. Various algorithms for automatic construction of parallel text such as STRAND, PTMiner, PTI, WPDE, BITS were studied to determine their appropriateness in creating English -Igala parallel text. Wikipedia and the Bible which are excellent sources of parallel or comparable corpora were also gleaned. Existing algorithms and other sources of Parallel text were found to be unsuitable for the construction of English -Igala parallel text due to the unavailability of contents rendered in Igala language on the web and in electronic form. A combination of manual and machine assisted translation was used to generate the parallel text. English -Igala parallel corpora comprising of 50,000 aligned sentences was obtained.

No abstract

No abstract