ULK Scientific Journal

Editorial

Claude Rusibana — 2024-10-10

No Abstract

Convolutional with attention gated recurrent network for sentiment analysis

Olivier Habimana — 2024-10-10

In recent years, deep learning approaches like convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have received much attention to natural language processing tasks, especially to sentiment analysis. Different methods will be used to measure the Convolutional with Attention Gated Recurrent Network for Sentiment Analysis. Thankfully, these methods achieved significant results. However, these approaches individually fail to accomplish the task of sentiment analysis at the extent level. In sentiment analysis, the likelihood of a given word is estimated based on long-term dependencies and local contextual features that depend on a word and its neighboring words. This paper suggests a Convolutional with Attention Gated Recurrent Network (CAGRN) model performs the sentiment analysis by extracting these features. The objective behind our model is to apply the CNN layer to extract local contextual features. Afterward, the CAGRNusesa bidirectional gated recurrent unit (Bi-GRU) layer to encode the long-term dependence features. On the other hand, the attention mechanism is applied to help our model select the convenient words that hold sentiment information. The CAGRN performs better in sentimentIn recent years, deep learning approaches like convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have received much attention to natural language processing tasks, especially to sentiment analysis. Different methods will be used to measure the Convolutional with Attention Gated Recurrent Network for Sentiment Analysis. Thankfully, these methods achieved significant results. However, these approaches individually fail to accomplish the task of sentiment analysis at the extent level. In sentiment analysis, the likelihood of a given word is estimated based on long-term dependencies and local contextual features that depend on a word and its neighboring words. This paper suggests a Convolutional with Attention Gated Recurrent Network (CAGRN) model performs the sentiment analysis by extracting these features. The objective behind our model is to apply the CNN layer to extract local contextual features. Afterward, the CAGRNusesa bidirectional gated recurrent unit (Bi- GRU) layer to encode the long-term dependence features. On the other hand, the attention mechanism is applied to help our model select the convenient words that hold sentiment information. The CAGRN performs better in sentiment analysis by using the learned features. Our approach achieves competitive results on two real datasets IMDb and SSTb, compared with baseline models and requires fewer parameters. Executing various ablation experiments of our model components will be done in future.

Electronic Voting System via GSM

Kabandana Innocent — 2024-10-10

Voting is a right for every citizen allowed to vote in democratic countries. Different countries are having manual or electronic systems to elect their Constitutional Law and their different leaders like President, Prime minister, Member of parliaments, senates, etc. Electronic Voting System via GSM will provide additional facilities to the voters and candidates, to make election more flexible and efficient compare to the traditional election. We will use Global System for Mobile Communication to facilitate the candidates and voters to use this technology to register, elect their candidates from their places and the total votes will be published in a very short time period. This system is capable to enhanced voter verification and mobility while maintaining voter privacy with One Time Password (OTP) generation. Our main objective is to design a secure GSM Mobile Communication Electronic Voting (GMC-EV) model to establish secured connectivity and One Time Password (OTP) based authorization during GSM based electoral process to enhance the authentication of the system. We will use Software Development life Cycle model as our methodology to implement our Electronic Voting System via GSM. The key findings is the test bed simulation of the proposed GSM-Electronic Voting System, the other key findings include the comparison of the time analysis of Secure Hash Algorithm 1(SHA-1) and Secure Hash Algorithm 2(SHA-2). Our system will based on GlobalMobile Communication Electronic Voting as different people are now having smart phones, One Time Password will be generated for every voter. As a recommendation, the Election Commission of Voting System should train the key parsons who will make the follow up, different activities of the elections from the beginning to the end of the election.

Replication of success for sustainable rural livelihood development with inclusive innovation in Rwanda: A case of neighbouring communities Inngeruka sector, Eastern Province

Jean de Dieu Mfitiryayo — 2024-10-10

The current paper intended to examine the replication of success for sustainable rural livelihoods development with inclusive innovation in Rwanda. It was undertaken under four specific objectives which involved to determine the features of replication of success among neighboring communities, to examine the types of channels for replication of success, to assess the extent to which the replication of success sustains the neighboring communities’ livelihoods and too determine the moderating effects of replication of success among neighboring communities in Ngeruka Sector. The data underlying this study were collected within10 months of fieldwork since November2020 to August 2021 in 12 villages of Ngeruka sector in Bugesera District- Eastern Province. Among 800 most vulnerable people, the 300 were randomly selected among others to participate in the study. The qualitative and quantitative data were collected using 300 community memberbased questionnaires, administered in the local language. The participants to whom the questionnaires were addressed corresponded to 293 while other 7 were interviewed. The 12 Focus Group Discussions were also conducted to 24 groups included the 12 pre-assisted groups while other 12 referred to the recipients of successes. In the absence of an official census, a household register was compiled consisting of 3,170 community members (in neighboring communities). It was seen that, this approach helped to reduce and combat conflict among households as resulted from poor living conditions (33.6%) and reduced the culture of dependence on external or foreign donors or to remain a burden on the government with 61.7% of respondents affirmed. The study explained on how the [79] mutual support among the communities themselves was a good way to keep them from migrating to other parts of the country for searching for lives, rather enabled them maintaining their community (63.6%).In addition, infrastructures and other natural resources were sustainably managed; for example, the people were able to use their resources wisely and make them productive such as land (76.6%), water with 3.6%, livestock (67.3%) and money with 64.3%. This approach identified itself as the only way to help communities cope with their problems and be able to find lasting solutions. Empowering communities as well as giving them opportunity to make decision have increased their motivation and reduce their resistance to community changes. The study encouraged public, private and non-governmental organizations to take advantages on the applicability of replication of success towards the community resilience and sustainable livelihoods development. All concerned organizations are recalled to help communities bearing the culture of ownership spirit as the real pathway for the replication of success to take place. The findings of this study are highly privileged to recommend the government of Rwanda to adopt the replication of success related policy among others.

An evaluation of financial literacy among business owners of micro, small and medium enterprises in Rwanda

Amani Manzi Alain — 2024-10-10

This paper is an evaluation of financial literacy among business owners of Micro, Small and Medium Enterprises in Rwanda. The objectives of the study were to determine: Levels of financial knowledge; levels of financial skills; nature of attitude towards financial matters and nature of financial behaviour. The study is hypothesized on the fact that a combination of high financial knowledge level, high financial skill level, good financial behaviour and positive attitude towards financial matters results in high level of financial literacy. The study employed a crosssectional survey design covering 200 participants. This sample was determined using Sloven’s formula. Data was collected using a questionnaire and analyzed using MS Excel. The key findings were that 72.7% of micro business owners have a high financial knowledge level, 77.7% have a high skill level, 69.3% have good attitude towards financial matters and 69.7% were practicing good financial behaviour. Also, 73.7% of small business owners have high financial knowledge, 79.5% are skilled in financial matters, 71.0% have good attitude towards financial matters and 76.2% were practicing good financial behaviour. Other findings are that 85.5% of the medium business owners have high financial knowledge level, 87.8% have high financial skill level, 73.7% have good attitude towards financial matters and 77.1% were exhibiting good financial behaviour. The study recommends that the government should identify the 13.5% of micro business owners, the 15.1% of small business owners and the 9.8% of medium business owners who are not knowledgeable and educated them on financial matters so that they also become knowledgeable; identify the 9.4% of micro business owners, 9.9% of small business owners and 6.4% of the medium business owners who are not skilled, and train them in appropriate financial skills so that they become adequately skilled; identify the 19.7% of the micro business owners, the 19.7% of small business owners and the 20.9% of the medium business owners who have poor attitude towards financial matters, and then sensitize them so as to change their mind set for them to adopt a good attitude; identify the 18.5% of the micro business owners, the 14.2% of the small business owners and the 16.4% of the medium business owners who practice bad financial behaviour and then educate them on good financial behaviour.