The current XAI techniques present explanations mainly as visuals and structured data. However, these explanations are difficult to interpret for a non-expert user. Here, the use of natural language generation (NLG)-based techniques can help to represent explanations in a human-understandable format. The paper addresses the issue of automatic generation of narratives using a modified transformer approach. Further, due to the unavailability of a relevant annotated dataset for development and testing, the authors also propose a verbalization template approach to generate the same. The input of the transformer is linearized to convert the data-to-text task into text-to-text task. The proposed work is evaluated on a verbalized explained PIMA Indians diabetes dataset and exhibits significant improvement as compared to existing baselines for both manual and automatic evaluation. Also, the narratives provide better comprehensibility to be trusted by human evaluators than the non-NLG counterparts. Lastly, an ablation study is performed in order to understand the contribution of each component.

The mobile ad hoc network (MANET) is a hostile environment to design and implement computing algorithms. In MANETs, the coordinator election is the opening step to execute any service. Thus, the leader election is, de facto, prerequisite to various applications e.g., agreement, inter node communication, data exchange, serve the incoming request, key distribution, group communication, grant privileges, etc. The message propagation is costly affair as compared to computation. The MANET being bandwidth and power scarce, the development of message efficient protocols has been a preferred problem of research. Hence, in the contemporary literature, several protocols exist for leader election which focuses on reducing the number of algorithmic messages. The present article proposes a message efficient and failure resilient technique that improves leader availability. The proposed technique elects a vice-coordinator and a group of elite nodes as cabinet that avoids frequent leader election in the event of leader crash. Unlike, most of the contemporary leader election protocols, it avoids the broadcast of messages, to a large extent, and uses multicast and eventually unicast in leader election. The illustration also includes the simulated results and efficiency analysis of the proposed technique.

The paper presents an authentication based Byzantine fault tolerant agreement protocol that is capable of detecting the tentative failures in transaction processing systems. The protocol employs a cryptographic hash function to detect crash as well as few more types of commonly faced arbitrary behaviors manifested by the participating server replicas in the system. We show that use of hash functions can successfully handle various types of (arbitrary) faults. In addition, it suggests a preemptive mechanism to proactively detect the malicious replicas. The protocol avoids frequent view changes that contributes to the liveness of the system and ensures continuity in transaction processing.

The cognitive radio network is an emerging wireless communication and computing paradigm where the network is not equipped with control infrastructure. The cognitive radios are allowed to execute the processes over the multiple heterogeneous channels without perturbing the licensed users of respective channels. The termination detection is a fundamental and non-trivial problem in the distributed computing systems. In this article, we present a termination detection protocol for heterogeneous multihop cognitive radio network. The proposed protocol applies the credit distribution and recovery approach. We present the two types of termination detection, i.e., strong and weak termination. Also, we give the correctness proof of the protocol.

This letter presents a new filtering scheme based on contrast enhancement within the filtering window for removing the random valued impulse noise. The application of a nonlinear function for increasing the difference between a noise-free and noisy pixels results in efficient detection of noisy pixels. As the performance of a filtering system, in general, depends on the number of iterations used, an effective stopping criterion based on noisy image characteristics to determine the number of iterations is also proposed. Extensive simulation results exhibit that the proposed method significantly outperforms many other well-known techniques.