3D Vision-Language Foundation Models (VLFMs) have demonstrated strong generalization and zero-shot recognition capabilities in open-world point cloud processing tasks. However, their performance often degrades in practical scenarios where data are noisy, incomplete, or drawn from distributions that differ from the training data. To address this challenge, we propose Uni-Adapter, a novel training-free online test-time adaptation (TTA) strategy for 3D VLFMs based on dynamic prototype learning. Uni-Adapter maintains a 3D cache that stores class-specific cluster centers as prototypes, which are continuously updated to capture intra-class variability under heterogeneous data distributions. These dynamic prototypes serve as anchors for cache-based logit computation through similarity scoring. In parallel, a graph-based label smoothing module models inter-prototype similarities to enforce label consistency among related prototypes. Finally, predictions from the original 3D VLFM and the refined 3D cache are unified through entropy-weighted aggregation to ensure reliable adaptation. Without retraining, Uni-Adapter effectively mitigates distribution shifts and achieves state-of-the-art performance across diverse 3D benchmarks and multiple 3D VLFMs, improving performance on ModelNet-40C by 10.55%, ScanObjectNN-C by 8.26%, and ShapeNet-C by 4.49% over the source 3D VLFMs.

Despite various supervised methods and algorithms, prediction methods do not necessarily provide the optimal values of the outputs. In this paper, an approach is proposed based on the integration of the clustering algorithm and a new mathematical programming model for predicting the efficient and optimal level of hospital outputs. In the first stage, units are evaluated using data envelopment analysis (DEA) and then, high-efficiency units are selected. In the second stage, the selected units are clustered based on the inputs of the units using the fuzzy C-means algorithm. For each unit to be estimated, the corresponding cluster is found and the closest unit to the target unit is determined. Finally, using the proposed new mathematical programming model, the optimal values are estimated based on the more efficient unit. A case study on Iranian hospitals illustrates the implementation of algorithms and methods and shows the abilities of the proposed approach.

The intention of target setting for Decision-Making Units (DMUs) in Data Envelopment Analysis (DEA) is to perform better than their peers or reach a reference efficiency level. However, most of the time, the logic behind the target setting is based on mathematical models, which are not achievable in practice. Besides, these models are based on decreasing/increasing inputs/outputs that might not be feasible based on DMU's potential in the real world. We propose a data-driven decision support framework to set actionable and feasible targets based on vital inputs-outputs for target setting. To do so, DMUs are classified in their corresponding Efficiency Frontier (EF) levels based on multiple EFs approach and a machine learning classifier. Then, the vital inputs-outputs are determined using an Explainable Artificial Intelligence (XAI) method. Finally, a Multi-Objective Counterfactual Explanation is developed based on DEA (MOCE-DEA) to lead DMU in reaching the reference EF by adjusting actionable and feasible inputs-outputs. We studied Iranian hospitals to evaluate the proposed framework and presented two cases to demonstrate its mechanism. The results show that the performance of the DMUs is improved to reach the reference EF for studied cases. Then, a validation was conducted with the primal DEA model to show the robust improvement of DMUs after adjusting their original value based on the generated solutions by the proposed framework. It demonstrates that the adjusted values can also improve DMUs' performance in the primal DEA model.

Proactive identification and the management of disruption risks play a crucial role in the achievement of a global supply chain’s aims. Given the velocity and volume by which such disruption events occur, it is impractical to expect supply chain managers to determine the occurrence of such events manually. Given the pressures facing global supply chains due to the COVID-19 crisis, it is important for supply chain managers to proactively identify disruption risks to their supply chains and manage them to either achieve the outcomes or develop plans by which resilience against them can be built. In this paper, we demonstrate how the integration of natural language processing and reinforcement learning, which are fundamental artificial intelligence methods, can be used to assist supply chain risk managers in the timely identification of such disruption events. We explain in detail our proposed approach, namely RL-SCRI and show its superiority over the current models in achieving its aim.

Financial institutions benefit from the advanced predictive performance of machine learning algorithms in automatic decision-making for credit scoring. However, two main challenges hamper machine learning algorithms’ applicability in practice: the complex and black-box nature of algorithms that hinder their understandability and the inability to guide rejected customers to have a successful application. Regarding customer relationship management is one of the main responsibilities of financial institutions; they must clarify the decision-making process to guide them. However, financial institutions are not willing to disclose their decision-making procedure to prevent potential risks from customers or competitors side. Hence, in this study, a decision support framework is proposed to clarify the decision-making process and model strategic decision-making to guide rejected customers simultaneously. To do so, after classifying customers in their corresponding groups, the capability of Shapley additive exPlanations method is exploited to extract the most impactful features to the prediction’s outcome globally and locally. Then, based on the benchmarking approach , the equivalent approved peer is found for the rejected customer for target setting to modify the application. To find the optimal modified values for a counterfactual prediction, a multi-objective gamed-based counterfactual explanation model is developed using the prisoner’s dilemma game as the constraint to simulate strategic decision-making. After optimization, the decision is reported to the customers concerning the credential background. A public data set is used to elaborate on the proposed framework. This framework can generate counterfactual predictions successfully by modifying perspective features.

Blockchain technology has the potential to be applied widely in supply chain operations. One such area is proactive supply chain risk management (SCRM). In this area, existing researchers have highlighted the fraudulent behaviour of supply chain partners who do not disclose information on the risks that impact their operations. Blockchain can address this problem by encoding each partner’s commitment to SCRM and achieve consensus. However, before this can be achieved, a key challenge to address is the inability of existing consensus mechanisms such as Proof of Work (PoW), Proof of Authority (PoA) and Proof of Stake (PoS) to deal with information that does not have a digital footprint. In this paper, we address this gap by proposing the Proof by Earnestness (PoE) consensus mechanism which accounts for the authenticity, legitimacy and trustworthiness of information that does not have a digital footprint. We also propose the Subjective Information Authenticity Earnestness Framework (SIAEF) as the overarching framework that assists PoE in achieving its aim. We test the applicability of SIAEF and PoE in a real-world blockchain environment by deploying it as a decentralized application (Dapp) and applying it in BscScan Testnet which is an official test blockchain network.

The accuracy and reliability of XAI methods are important to establish their credibility and use in complex decision-making tasks. Existing XAI methods provide little information about the correctness and reliability of their outputs. Furthermore, post-hoc explanation approaches explain the outcomes after producing them, not in a step-by-step glass-box manner to explain how an output is reached. Our proposed approach addresses these drawbacks by designing a Belief-Rule-Based (BRB) framework that interprets in a glass-box manner why a particular decision has been reached. It does that by determining the chance of different output classes occurring for a specific time period by considering the different possible permutations of the inputs along with their influence. This also assists the user to determine if the given input dataset is incomplete, vague, imprecise or inconsistent before trusting the analysis emanating from it. We compare the performance of the proposed BRB approach against the different eXplainable artificial intelligence (XAI) methods, such as SHAP, LIME and LINDA-BN to ensure the users of the trustworthiness of its analysis. This also enables users to determine the extent to which each of the XAI techniques meets the requirements of XAI and the gaps that need to be addressed.

Abstract Skin cancer, primarily resulting from the abnormal growth of skin cells, is among the most common cancer types. In recent decades, the incidence of skin cancer cases worldwide has risen significantly (one in every three newly diagnosed cancer cases is a skin cancer). Such an increase can be attributed to changes in our social and lifestyle habits coupled with devastating man-made alterations to the global ecosystem. Despite such a notable increase, diagnosis of skin cancer is still challenging, which becomes critical as its early detection is crucial for increasing the overall survival rate. This calls for advancements of innovative computer-aided systems to assist medical experts with their decision making. In this context, there has been a recent surge of interest in machine learning (ML), in particular, deep neural networks (DNNs), to provide complementary assistance to expert physicians. While DNNs have a high processing capacity far beyond that of human experts, their outputs are deterministic, i.e., providing estimates without prediction confidence. Therefore, it is of paramount importance to develop DNNs with uncertainty-awareness to provide confidence in their predictions. Monte Carlo dropout (MCD) is vastly used for uncertainty quantification; however, MCD suffers from overconfidence and being miss calibrated. In this paper, we use MCD algorithm to develop an uncertainty-aware DNN that assigns high predictive entropy to erroneous predictions and enable the model to optimize the hyper-parameters during training, which leads to more accurate uncertainty quantification. We use two synthetic (two moons and blobs) and a real dataset (skin cancer) to validate our algorithm. Our experiments on these datasets prove effectiveness of our approach in quantifying reliable uncertainty. Our method achieved 85.65 ± 0.18 prediction accuracy, 83.03 ± 0.25 uncertainty accuracy, and 1.93 ± 0.3 expected calibration error outperforming vanilla MCD and MCD with loss enhanced based on predicted entropy.

Recommender systems have seen wide adoption in different domains. The motive of such systems has evolved from providing generic recommendations in the past to providing customized and user-focused recommendations. To achieve this aim, the complexity and sophistication of the underlying techniques such systems use have evolved. Current recommender systems use advanced Artificial Intelligence techniques to provide intelligent recommendations and adapt their future workings to the user’s interest and requirements. One such technique currently being used in the literature to achieve this aim is Reinforcement Learning. However, a drawback of this technique is that it is data intensive and needs to be trained on data that represent different scenarios to ensure that the recommended output in a given scenario is accurate. In this article, we present an approach, namely Reinforcement Learning-based News Recommendation System (RL-NRS), to address this drawback in the domain of news recommendation. We explain the different stages of RL-NRS in detail and compare its performance with news articles recommended by Google for a particular search term.

As the volume of scientific papers grows rapidly in size, knowledge management for scientific publications is greatly needed. Information extraction and knowledge fusion techniques have been proposed to obtain information from scholarly publications and build knowledge repositories. However, retrieving the knowledge of problem/solution from academic papers to support users on solving specific research problems is rarely seen in the state of the art. Therefore, to remedy this gap, a knowledge-driven solution support system (K3S) is proposed in this paper to extract the information of research problems and proposed solutions from academic papers, and integrate them into knowledge maps. With the bibliometric information of the papers, K3S is capable of providing recommended solutions for any extracted problems. The subject of intrusion detection is chosen for demonstration in which required information is extracted with high accuracy, a knowledge map is constructed properly, and solutions to address intrusion problems are recommended.

Assortment planning is one of the fundamental and complex decisions for online retailers. The complexity of this problem is increasing while considering demand and supply uncertainties in assortment planning (AP). However, this leads to more efficient results in today's uncertain markets. In this paper, the supplier and E-tailer interactions are modeled by the non-cooperative game theory model. As small-scale online retailers opposed to bricks and mortar usually have lower power in front of suppliers, we propose a Stackelberg or leader-follower game model. First, the supplier as a leader announces its decisions regarding selling price to the E-tailer. Consequently, the E-tailer reacts by determining the purchase quantity, selling price to the customers and assortment size. Various scenarios are presented and analyzed to show the effectiveness of the Stackelberg game model in simulating the interactions between small-scale online retailers and a powerful supplier.

We propose an interactive decision-making framework to assist a Customer Service Representative (CSR) in the efficient and effective recognition of customer records in a database with many ambiguous entries. Our proposed framework consists of three integrated modules. The first module focuses on the detection and resolution of duplicate records to improve effectiveness and efficiency in customer recognition. The second module determines the level of ambiguity in recognizing an individual customer when there are multiple records with the same name. The third module recommends the series of feature-related questions that the CSR should ask the customer to enable rapid recognition, based on that level of ambiguity. In the first module, the F-Swoosh approach for duplicate detection is used, and in the second module a dynamic programming-based technique is used to determine the level of ambiguity within the customer database for a given name. In the third module, Levenshtein edit distance is used for feature selection in combination with weights based on the Inverse Document Frequency (IDF) of terms. The algorithm that requires the minimum number of questions to be put to the customer to achieve recognition is the algorithm that is chosen. We evaluate the proposed framework on a synthetic dataset and demonstrate how it assists the CSR to rapidly recognize the correct customer.

The concept of stratification (CST) is a reform in problem solving approaches in computer science introduced and developed by Lotfi Zadeh [1]. In this approach, the target set is a set of the initial states. The environment around the target is then systematically identified and strata around the target are gradually built using the concept of `incremental target enlargement'. We advocate that this approach is useful in different areas; however, it is essential to highlight the potential applicability of the approach by presenting different examples. In this paper illustrative examples in Information Dominance (ID) and in requirement elicitation in contracting, are structured using CST. These examples show the application of the concept in the contexts of logistic informatics and contracting and assists the modelling process. Of particular relevance are Example 4 and 5. Example 4 exposes situations that are not dealt with in the current form of CST. The study, therefore, extends CST by considering the possibility of occasional state repetition while using the same input. Further, a 3D version of CST is structured and examined through example 5. This proposed version of the concept illustrates how the Fuzzy Inference System (FIS) benefits the user when applying CST.

Health, Safety, Environment and Ergonomie (HSEE) are important factors in any organization. An organization always have to assess its compliance in these factors to the required benchmarks and take proactive actions to improve them if required. In this paper, we propose a Fuzzy Cognitive Map-Bayesian network (BN) model in order to assist organizations in doing this process. Fuzzy Cognitive Map (FCM) method is used for constructing graphical model of BN to ascertain the relationships between the inputs and the impact which they will have on the quantified HSEE. Noisy-OR method and EM are used to ascertain the conditional probability between the inputs and quantifying the HSEE value. Using this, we find out the most influential input factor on HSEE quantification which can then be managed for improving an organization's compliance to HSEE. Leveraging the power of Bayesian network in modeling HSEE and augmenting it with FCM is the main contribution of this research work which opens this line of research.

Avoiding ambiguity and fuzziness in the determination of the requirements is a crucial factor in the success of Performance Based Contracting (PBC). To date, there is a research gap because insufficient studies have been undertaken to address this significant issue in the pro-curement process. Previous studies that have been con-ducted on requirement specification and elicitation are limited to software engineering. This study investigates this issue in the procurement process and proposes an integrated framework using Natural Language Pro-cessing (NLP), game theory and fuzzy logic. This re-search contributes to contract theory by opening a new line of research which paves the way for leveraging arti-ficial intelligence techniques in automated or semi-automated contract monitoring.

The big data research topic has grown rapidly for the past decade due to the advent of the "data deluge". Recent advancements in the literature leverage human computing power known as crowdsourcing to manage and harness big data for various applications. However, human involvement in the completion of crowdsourcing tasks is an error-prone process that affects the overall performance of the crowd. Thus, controlling the quality of workers is an essential step for crowdsourcing systems, which due to unavailability of ground-truth data for any task at hand becomes increasingly challenging. To propose a solution to this problem, in this study, we propose OSQC (Online Statistical Quality Control Framework) for managing the performance of workers in crowdsourcing. OSQC ascertains the worker's performance by using a statistical model and then leverages the traditional statistical control techniques to decide whether to retain a worker for crowdsourcing or to evict him. We evaluate our proposed framework on a real dataset and demonstrate how OSQC assists crowdsourcing to maintain its accuracy.

Building ontology for wireless network intrusion detection is an emerging method for the purpose of achieving high accuracy, comprehensive coverage, self-organization and flexibility for network security. In this paper, we leverage the power of Natural Language Processing (NLP) and Crowdsourcing for this purpose by constructing lightweight semi-automatic ontology learning framework which aims at developing a semantic-based solution-oriented intrusion detection knowledge map using documents from Scopus. Our proposed framework uses NLP as its automatic component and Crowdsourcing is applied for the semi part. The main intention of applying both NLP and Crowdsourcing is to develop a semi-automatic ontology learning method in which NLP is used to extract and connect useful concepts while in uncertain cases human power is leveraged for verification. This heuristic method shows a theoretical contribution in terms of lightweight and timesaving ontology learning model as well as practical value by providing solutions for detecting different types of intrusions.

Trust modeling of both the interacting parties in a virtual world, is a critical element of business intelligence. A key aspect in trust modeling is to be able to accurately predict the future trust value of an interacting party. In this paper, we propose an intelligent method for predicting the future trust value of a trusted entity. We propose the use of Z-number to represent both the trust value and its corresponding reliability. Subsequently, we apply Artificial Neural Network (ANN) to predict future trust values. We generate a large number of synthetic time series, with a view to model real-world trust values of trusted entity. We validate the working of our methodology using the generated time series.

In today's highly dynamic economy and society, the performance evaluation of Decision Making Units (DMUs) is of high importance. This study presents an efficient model for analyzing the outputs of performance measurement methodologies by means of trust, which provides explicit qualitative scales instead of representing pure numerical data. The efficiency rate of the current, previous and coming years, as well as the average efficiency and standard deviation, are the five inputs for this model. These efficiency rates are calculated using Data Envelopment Analysis (DEA). The approach uses time series forecasting to predict the future efficiency rate. Furthermore, the implemented Auto Regressive (AR) model includes an Auto Correlation Function (ACF) for input selection. The model utilizes T-norms and S-norms as the final modeling tools. To illustrate the applicability of the proposed model, we apply it to a data set of DMUs. Ultimately, modified trust values for these DMUs are determined using the proposed approach.