To address the challenge of requiring extensive data and computational resources in intelligent control game systems, this paper proposes a triangular game relationship model on a plane, involving dynamic player A, dynamic player B, and the fixed-point base camp b of player B. The model, which is typically used in real game scenarios where only small-scale effective data can be obtained to support the operation of intelligent game control systems, eliminates the necessary need for complete information in the adversarial deterrence game. In this triangular game relationship model, the coordinate position relationships between player A, player B, and player B's base camp b provide the constraint conditions for optimizing confrontation positions. Subsequently, an objective function for player A's optimal game position is constructed, and the local optimal confrontation position is derived and verified by incorporating the optimization constraints.
Qualitative models help policymakers understand land-use changes by examining the processes involved and predicting future patterns. This study aims to use dynamic synthesis methodology to establish cause-and-effect relationships. The study, conducted in the Mt. Elgon region encompassing Mbale, Manafwa, Bududa, and Namisindwa districts, utilized questionnaires and interviews to gather data. Findings suggest that short-term solutions to reduce wood fuel consumption include raising awareness, promoting mindset change, and providing subsidies for alternative energy sources. Long-term strategies involve advocating for tree planting, strengthening land use policies, and ensuring strict implementation. While stakeholder engagement does not automatically guarantee high-quality decision analysis, it is crucial for promoting natural resource protection and fostering a sense of ownership over the issue.
This paper focuses on using the cooperative neuro-fuzzy system for the effective and customised selection of entities from large and heterogeneous resources by presenting a general architecture. An experiment is carried out with the fast-moving consumer goods to prove the utility of the architecture. It is observed that most consumers go for the frequent purchase of fast-moving consumer items. Further, various brands, costs, discounts, schemes, quantities, and reviews might make it challenging. Hence, such decisions need to be intelligent and practically feasible in terms of time and effort. The paper discusses neural networks to categorise the entities, type-1 & 2 fuzzy membership functions with rules, training sets, and graphical views of the fuzzy rules and the experiment details. Besides the generic approach and experiment, the paper also discusses the work done so far with their limitations and applications in other domains. At the end, the paper presents the limitations and possible future enhancements.
In the recent times transfer learning models have known to exhibited good results in the area of text classification for question-answering, summarization, next word prediction but these learning models have not been extensively used for the problem of hate speech detection yet. We anticipate that these networks may give better results in another task of text classification i.e. hate speech detection. This paper introduces a novel method of hate speech detection based on the concept of attention networks using the BERT attention model. We have conducted exhaustive experiments and evaluation over publicly available datasets using various evaluation metrics (precision, recall and F1 score). We show that our model outperforms all the state-of-the-art methods by almost 4%. We have also discussed in detail the technical challenges faced during the implementation of the proposed model.
Ballots are often hold for fair decisions such as party theme selecting, however, the existing traditional ballot has some problems involving amount of human resources, cost of places, equipment, time and traffic, and repeated procedures. In order to solve the issues aforementioned, a ballot blockchain system is designed and implemented based on the smart contract of Ethereum. It is designed on the core blockchain technologies of the decentralized ledger technology, using a secure hash algorithm, anonymous user, incorruptible data, and adopting a public blockchain. The ballot blockchain system is implemented based on the MetaMask verification and the Remix interface development environment. The smart contract plays the role of the decision-maker for controlling ballot activities instead of numerous human tasks. All ballot transactions are recorded in the ballot blockchain permanently when the ballot completed. The aim of the ballot blockchain system is to achieve a fair, less time-consuming, secured, and transparent environment.
