Local Interpretable Model-agnostic Explanations Research Articles

Human addictive behavior prediction by using lime with ensemble model

<span>The data-driven techniques have utilized data mining and machine learning (ML) techniques in the biomedical and healthcare fields. The process of decision-making in uncertain contextual related to human addictions and emotions play an important role in the present research. The main aim of the research is to perform classification and generate a support system for uncertain addiction circumstances by proposing a technique for drug addiction treatment. The human behavior has majority shown challenges for the prediction of human behaviors that includes body poses estimation, movements and interaction with objects. This pose estimation has showed complexity with more pose aspects and the proposed research attempts to understand the human behaviors. The present research uses the local interpretable model-agnostic explanations (LIME) for finding the input features which are most important to generate a particular output based on decision service. LIME understands the model to perturb the data samples as an input and understands shows predictions change. Also, the ensemble classifier contains classifiers group that combines for performing the prediction of all unseen instances based on voting. The proposed LIME Feature-Ensemble classifier obtained 97.54% of accuracy when compared to the existing convolutional neural network (CNN) of 59.33% and Ensemble model of 93.33% accuracy.</span>

Wearable sensor devices can automatically identify the ON-OFF status of patients with Parkinson's disease through an interpretable machine learning model.

Accurately and objectively quantifying the clinical features of Parkinson's disease (PD) is crucial for assisting in diagnosis and guiding the formulation of treatment plans. Therefore, based on the data on multi-site motor features, this study aimed to develop an interpretable machine learning (ML) model for classifying the "OFF" and "ON" status of patients with PD, as well as to explore the motor features that are most associated with changes in clinical symptoms. We employed a support vector machine with a recursive feature elimination (SVM-RFE) algorithm to select promising motion features. Subsequently, 12 ML models were constructed based on these features, and we identified the model with the best classification performance. Then, we used the SHapley Additive exPlanations (SHAP) and the Local Interpretable Model agnostic Explanations (LIME) methods to explain the model and rank the importance of those motor features. A total of 96 patients were finally included in this study. The naive Bayes (NB) model had the highest classification performance (AUC = 0.956; sensitivity = 0.8947, 95% CI 0.6686-0.9870; accuracy = 0.8421, 95% CI 0.6875-0.9398). Based on the NB model, we analyzed the importance of eight motor features toward the classification results using the SHAP algorithm. The Gait: range of motion (RoM) Shank left (L) (degrees) [Mean] might be the most important motor feature for all classification horizons. The symptoms of PD could be objectively quantified. By utilizing suitable motor features to construct ML models, it became possible to intelligently identify whether patients with PD were in the "ON" or "OFF" status. The variations in these motor features were significantly correlated with improvement rates in patients' quality of life. In the future, they might act as objective digital biomarkers to elucidate the changes in symptoms observed in patients with PD and might be used to assist in the diagnosis and treatment of patients with PD.

Power Consumption and Processing Time Estimation of CNC Machines Using Explainable Artificial Intelligence (XAI)

Due to environmental issues such as climate change, companies are required to optimize their resource and energy consumption in their production process. Predicting power consumption and processing time of all production facilities is essential for manufacturing to develop mechanisms to prevent energy and resource waste and optimize their use. Machine learning is a powerful tool for prediction tasks using data in digitalized environments. In this paper, we present power consumption and processing time prediction of CNC milling machines using five machine learning regression models, i.e., decision tree, random forest, support vector regression (SVR), extreme gradient boosting (XGBoost), and artificial neural network (ANN). Since most of those models are black-box, we applied two explainable artificial intelligence (XAI) approaches, SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME), to give post-hoc explanations of the predictions given by the machine learning models. Our experiments indicated that random forest regression performed the best in predicting power consumption and processing time. The explanation showed that the number of axis rotations and the number of travels to the machine's zero point in rapid traverse were the most important factors that affected the processing time and power consumption. The companies using CNC milling machines can use our prediction models to optimally plan and schedule the operation of the milling machines in a time and energy-efficient manner. They can also optimize the factors that affect power consumption and processing time the most.

A Salp Swarm Algorithm for Interpreting Model Predictions

The Internet of Things (IoT), is changing practically every aspect of modern life. The proliferation of IoT has led to a rise in the frequency of cyber catastrophes. The threat landscape that security professionals face is dynamic, complex, and diversified. This paper proposes a novel approach to enhance Internet of Things applications by fusing the swarm intelligence of Salp Swarm Algorithms (SSA) with the predictive power of Random Forest (RF) and Decision Tree (DT) models Even though there is a lot of interest in the topic of explainable Artificial Intelligence (XAI) these days, more research is still needed to fully understand how successful XAI is at finding attack surfaces and vectors when implemented in cyber security applications. The growing use of machine/deep learning models in cyber defense, especially anomaly-based IDS, requires understanding the architecture of the models and providing evidence for their predictions to determine the probability of intrusions. Numerous approaches to interpretation have been proposed. They help researchers comprehend things like which variables have influenced the machine learning predictions. In this paper, we primarily address two popular local interpretation methods in machine learning: Shapley values and Local Interpretable Model-Agnostic Explanations (LIME).

Neutrosophic Meta SHAP and Neutrosophic Meta LIME: An Efficient Framework for Explainable AI in Oral Cancer Detection

Among the current generation researcher, artificial intelligence has played vital role in various fields, including healthcare. One of the key areas where it has shown enormous potential is in cancer detection and treatment. AI and methods of machine learning algorithms have been applied to analyze large datasets, such as genomics, transcriptomic, and imaging data, to identify patterns and relationships that can help in cancer diagnosis and therapy. However, due to the inherent complexity and heterogeneity of tumors in individual patients, building a diagnostic and therapeutic platform that can accurately analyze outputs becomes a challenging task. To address this challenge, researchers have proposed the use of explainable AI frameworks in cancer detection. Explainable AI frameworks aim to provide transparency and comprehensibility to the decision-making process of AI algorithms, ensuring that the predictions or classifications generated by these algorithms can be understood and trusted by healthcare professionals. One popular explainable AI method is SHAP (SHapley Additive explanations). SHAP is a well-known XAI method that provides intuitive and interpretable feature importance [13] for individual predictions. Another explainable AI method is LIME (Local Interpretable Model-agnostic Explanations), which generates posthoc explanations and is suitable for quick and satisfactory explanations. These existing explainable AI methods, however, have limitations in their applicability to cancer detection. Therefore, in this research article, we propose the use of two novel frameworks: Neutrosophic Meta SHAP and Neutrosophic Meta Lime. Neutrosophic Meta SHAP and Neutrosophic Meta Lime are efficient frameworks specifically designed for the analysis and interpretation of AI models in oral cancer detection.

Explainable AI in Deep Learning Based Classification of Fetal Ultrasound Image Planes

Fetal ultrasound images are widely used for visualizing fetal development during pregnancy. These ultrasound image planes provide information about the anatomy of the fetus, thus helping healthcare professionals identify any abnormalities. Several AI tools are now being applied to classify fetal planes automatically. Accurate classification of fetal ultrasound image planes is crucial for the correct prenatal diagnosis and healthcare. However, while deep learning models have shown promise in image classifications, their ”black box” nature makes their decisions challenging to interpret, which is a significant concern in healthcare analytics. This paper addresses the problem of interpretability of the decisions made by a Convolutional Neural Network(CNN) for fetal ultrasound image classification using an XAI technique. Despite their accuracy, the established solutions lack the transparency essential for medical professionals to trust the model's predictions. The LIME(Local Interpretable Model-agnostic Explanations) is applied to interpret the CNN classification that provides a high classification accuracy. The results of LIME interpretability model on the top of CNN highlight the critical regions that positively and negatively contribute to the classification decision. This approach offers a transparent and trusted solution for leveraging AI in prenatal diagnostics.

Employing Co-Learning to Evaluate the Explainability of Multimodal Sentiment Analysis

Deep neural nets are opaque black-box models with little to no understanding of underlying model dynamics. This issue is more prevalent in the case of multimodal artificial intelligence (AI) systems, where model explainability and interpretability are prime concerns due to data integration from heterogeneous data streams and complex inter and intramodal interactions. However, the traditional explainable models are challenging to apply in the multimodal scenario. We propose a co-learning-based solution for fostering model explainability for the natural language processing (NLP)-based multimodal sentiment analysis application to address this issue. The proposed approach employs explainability by obeying the co-learning principles of dealing with noisy and missing modality either at train or test time to find the modality dominance by extracting the local and global model explanations. The proposed approach is validated with post hoc explainability methods such as local interpretable model-agnostic explanations (LIME) and SHapley Additive exPlanations (SHAP) gradient-based explanations to model the modality contributions and interactions at the fusion level. The co-learning-based system ensures trust and robustness in the model by providing some degree of model explainability along with robustness. The kind of explanations provided is multifaceted and is obtained through a peek inside the black box, hence is specifically helpful for the system designers and model developers to understand the complex model dynamics that are far more challenging in the case of multimodal applications.

XAI Driven Intelligent IoMT Secure Data Management Framework.

The Internet of Medical Things (IoMT) has transformed traditional healthcare systems by enabling real-time monitoring, remote diagnostics, and data-driven treatment. However, security and privacy remain significant concerns for IoMT adoption due to the sensitive nature of medical data. Therefore, we propose an integrated framework leveraging blockchain and explainable artificial intelligence (XAI) to enable secure, intelligent, and transparent management of IoMT data. First, the traceability and tamper-proof of blockchain are used to realize the secure transaction of IoMT data, transforming the secure transaction of IoMT data into a two-stage Stackelberg game. The dual-chain architecture is used to ensure the security and privacy protection of the transaction. The main-chain manages regular IoMT data transactions, while the side-chain deals with data trading activities aimed at resale. Simultaneously, the perceptual hash technology is used to realize data rights confirmation, which maximally protects the rights and interests of each participant in the transaction. Subsequently, medical time-series data is modeled using bidirectional simple recurrent units to detect anomalies and cyberthreats accurately while overcoming vanishing gradients. Lastly, an adversarial sample generation method based on local interpretable model-agnostic explanations is provided to evaluate, secure, and improve the anomaly detection model, as well as to make it more explainable and resilient to possible adversarial attacks. Simulation results are provided to illustrate the high performance of the integrated secure data management framework leveraging blockchain and XAI, compared with the benchmarks.

A Machine Learning Explainability Tutorial for Atmospheric Sciences

Abstract With increasing interest in explaining machine learning (ML) models, this paper synthesizes many topics related to ML explainability. We distinguish explainability from interpretability, local from global explainability, and feature importance versus feature relevance. We demonstrate and visualize different explanation methods, how to interpret them, and provide a complete Python package (scikit-explain) to allow future researchers and model developers to explore these explainability methods. The explainability methods include Shapley additive explanations (SHAP), Shapley additive global explanation (SAGE), and accumulated local effects (ALE). Our focus is primarily on Shapley-based techniques, which serve as a unifying framework for various existing methods to enhance model explainability. For example, SHAP unifies methods like local interpretable model-agnostic explanations (LIME) and tree interpreter for local explainability, while SAGE unifies the different variations of permutation importance for global explainability. We provide a short tutorial for explaining ML models using three disparate datasets: a convection-allowing model dataset for severe weather prediction, a nowcasting dataset for subfreezing road surface prediction, and satellite-based data for lightning prediction. In addition, we showcase the adverse effects that correlated features can have on the explainability of a model. Finally, we demonstrate the notion of evaluating model impacts of feature groups instead of individual features. Evaluating the feature groups mitigates the impacts of feature correlations and can provide a more holistic understanding of the model. All code, models, and data used in this study are freely available to accelerate the adoption of machine learning explainability in the atmospheric and other environmental sciences.

Explainable AI for binary and multi-class classification of leukemia using a modified transfer learning ensemble model

Abstract In leukemia diagnosis, automating the process of decision-making can reduce the impact of individual pathologists' expertise. While deep learning models have demonstrated promise in disease diagnosis, combining them can yield superior results. This research introduces an ensemble model that merges two pre-trained deep learning models, namely, VGG-16 and Inception, using transfer learning. It aims to accurately classify leukemia subtypes using real and standard dataset images, focusing on interpretability. Therefore, the use of Local Interpretable Model-Agnostic Explanations (LIME) is employed to achieve interpretability. The ensemble model achieves an accuracy of 83.33% in binary classification, outperforming individual models. In multi-class classification, VGG-16 and Inception reach accuracies of 83.335% and 93.33%, respectively, while the ensemble model reaches an accuracy of 100%.

A Multi-objective Genetic Algorithm to Evolving Local Interpretable Model-agnostic Explanations for Deep Neural Networks in Image Classification

Deep convolutional neural networks have become a dominant solution for numerous image classification tasks. However, a main criticism is the poor explainability due to the black-box characteristic, which hurdles the extensive usage of deep convolutional neural networks. To address this issue, this paper proposes a new evolutionary multi-objective based method, which aims to explain the behaviours of deep convolutional neural networks by evolving local explanations on specific images. To the best of our knowledge, this is the first evolutionary multi-objective method to evolve local explanations. The proposed method is model-agnostic, i.e. it is applicable to explain any deep convolutional neural networks. ImageNet is used to examine the effectiveness of the proposed method. Three well-known deep convolutional neural networks -VGGNet, ResNet, and MobileNet, are chosen to demonstrate the modelagnostic characteristic. Based on the experimental results, it can be observed that the local explanations are understandable to end-users, who need to check the sensibility of the evolved explanations to decide whether to trust the predictions made by the deep convolutional neural networks. Furthermore, the local explanations evolved by the proposed method improves the confidence of deep convolutional neural networks making the predictions. Lastly, the pareto front and convergence analyses indicate that the proposed method can form a good set of nondominated solutions.

Applying local interpretable model-agnostic explanations to identify substructures that are responsible for mutagenicity of chemical compounds

“The study identifies more specific chemical fragments responsible for mutagenicity in human beings, aiding in chemical safety evaluation, and health and environmental risk assessment”.

Novel Approach to Intrusion Detection: Introducing GAN-MSCNN-BILSTM with LIME Predictions

This paper introduces an innovative intrusion detection system that harnesses Generative Adversarial Networks (GANs), Multi-Scale Convolutional Neural Networks (MSCNNs), and Bidirectional Long Short-Term Memory (BiLSTM) networks, supplemented by Local Interpretable Model-Agnostic Explanations (LIME) for interpretability. Employing a GAN, the system generates realistic network traffic data, encompassing both normal and attack patterns. This synthesized data is then fed into an MSCNN-BiLSTM architecture for intrusion detection. The MSCNN layer extracts features from the network traffic data at different scales, while the BiLSTM layer captures temporal dependencies within the traffic sequences. Integration of LIME allows for explaining the model's decisions. Evaluation on the Hogzilla dataset, a standard benchmark, showcases an impressive accuracy of 99.16% for multi-class classification and 99.10% for binary classification, while ensuring interpretability through LIME. This fusion of deep learning and interpretability presents a promising avenue for enhancing intrusion detection systems by improving transparency and decision support in network security.

Predictive modeling of consumer purchase behavior on social media: Integrating theory of planned behavior and machine learning for actionable insights.

In recent times, it has been observed that social media exerts a favorable influence on consumer purchasing behavior. Many organizations are adopting the utilization of social media platforms as a means to promote products and services. Hence, it is crucial for enterprises to understand the consumer buying behavior in order to thrive. This article presents a novel approach that combines the theory of planned behavior (TPB) with machine learning techniques to develop accurate predictive models for consumer purchase behavior. This study examines three distinct factors of the theory of planned behavior (attitude, social norm, and perceived behavioral control) that provide insights into the primary determinants influencing online purchasing behavior. A total of eight machine learning algorithms, namely K-nearest neighbor, Decision Tree, Random Forest, Logistic Regression, Naive Bayes, Support Vector Machine, AdaBoost, and Gradient Boosting, were utilized in order to forecast consumer purchasing behavior. Empirical findings indicate that gradient boosting demonstrates superior performance in predicting customer buying behavior, with an accuracy rate of 0.91 and a macro F1 score of 0.91. This holds true when all factors, namely attitude (ATTD), social norm (SN), and perceived behavioral control (PBC), are included in the analysis. Furthermore, we incorporated Explainable AI (XAI), specifically LIME (Local Interpretable Model-Agnostic Explanations), to elucidate how the best machine learning model (i.e. gradient boosting) makes its prediction. The findings indicate that LIME has demonstrated a high level of confidence in accurately predicting the influence of low and high behavior. The outcome presented in this article has several implications. For instance, this article presents a novel way to combine the theory of planned behavior with machine learning techniques in order to predict consumer purchase behavior. This integration allows for a comprehensive analysis of factors influencing online purchasing decisions. Also, the incorporation of Explainable AI enhances the transparency and interpretability of the model. This feature is valuable for organizations seeking insights into factors driving predictions and the reasons behind certain outcomes. Moreover, these observations have the potential to offer valuable insights for businesses in customizing their marketing strategies to align with these influential factors.

Enhancing explainability in brain tumor detection: A novel DeepEBTDNet model with LIME on MRI images

AbstractEarly detection of brain tumors is vital for improving patient survival rates, yet the manual analysis of the extensive 3D MRI images can be error‐prone and time‐consuming. This study introduces the Deep Explainable Brain Tumor Deep Network (DeepEBTDNet), a novel deep learning model for binary classification of brain MRIs as tumorous or normal. Employing sub‐image dualistic histogram equalization (DSIHE) for enhanced image quality, DeepEBTDNet utilizes 12 convolutional layers with leaky ReLU (LReLU) activation for feature extraction, followed by a fully connected classification layer. Transparency and interpretability are emphasized through the application of the Local Interpretable Model‐Agnostic Explanations (LIME) method to explain model predictions. Results demonstrate DeepEBTDNet's efficacy in brain tumor detection, even across datasets, achieving a validation accuracy of 98.96% and testing accuracy of 94.0%. This study underscores the importance of explainable AI in healthcare, facilitating precise diagnoses and transparent decision‐making for early brain tumor identification and improved patient outcomes.

Predictive modeling of perioperative blood transfusion in lumbar posterior interbody fusion using machine learning.

Background: Accurate estimation of perioperative blood transfusion risk in lumbar posterior interbody fusion is essential to reduce the number, cost, and complications associated with blood transfusions. Machine learning algorithms have the potential to outperform traditional prediction methods in predicting perioperative blood transfusion. This study aimed to construct a machine learning-based perioperative transfusion risk prediction model for lumbar posterior interbody fusion in order to improve the efficacy of surgical decision-making. Methods: We retrospectively collected clinical data on 1905 patients who underwent lumbar posterior interbody fusion surgery at the Second Hospital of Shanxi Medical University between January 2021 and March 2023. All the data was randomly divided into a training set and a validation set, and the "feature_importances" method provided by eXtreme Gradient Boosting (XGBoost) algorithm was applied to select statistically significant features on the training set to establish five machine learning prediction models. The optimal model was identified by utilizing the area under the curve (AUC) and the probability calibration curve on the validation set. Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME) were employed for interpretable analysis of the optimal model. Results: In the postoperative outcomes of patients, the number of hospital days in the transfusion group was longer than that in the non-transfusion group. Additionally, the transfusion group experienced higher total hospital costs, 90-day readmission rates, and complication rates within 90days after surgery than the non-transfusion group. A total of 9 features were selected for the models. The XGBoost model performed best with an AUC value of 0.958. The SHAP values showed that intraoperative blood loss, intraoperative fluid infusion, and number of fused segments were the top 3 most important features affecting perioperative blood transfusion in lumbar posterior interbody fusion. The LIME algorithm was used to interpret the individualized prediction. Conclusion: Surgery, ASA class, levels fused, total intraoperative blood loss, operative time, and preoperative Hb are viable predictors of perioperative blood transfusion in lumbar posterior interbody fusion. The XGBoost model has demonstrated superior predictive efficacy compared to the traditional logistic regression model, making it a more effective decision-making tool for perioperative blood transfusion.

Unlocking the Potential of CT scans: An Explanation-Driven Deep Learning Model for Predicting Lung Cancer

This research study aims to evaluate the effectiveness of transfer learning with the ResNet50 model for classifying CT scan images of lungs as having cancer or not. Additionally, it explores the interpretability methods of LIME (Local Interpretable Model-Agnostic Explanations) and SHAP (SHapley Additive exPlanations) to provide explanations for the predictions made by the ResNet50 model on CT scan images. The research objectives include developing a deep learning model based on the ResNet50 architecture, evaluating its performance using various metrics, and explaining the predictions using LIME and SHAP techniques. The dataset consists of a collection of CT scan images of lungs, with labels indicating the presence or absence of cancer. Through k-fold cross-validation, the model achieves high accuracy and low loss, demonstrating its effectiveness in classifying lung cancer. The interpretability methods of LIME and SHAP shed light on the crucial features and regions in the CT scan images that contribute to the model's predictions, enhancing the understanding of the model's decision-making process. The results highlight the potential of transfer learning and interpretability techniques in improving the accuracy and explainability of lung cancer detection models. Future directions may involve applying the developed model to larger datasets, classifying different stages of cancer, and identifying the specific regions within the lungs where cancer cells are detected.

Explainable Soft Attentive EfficientNet for breast cancer classification in histopathological images

Breast Cancer (BC) is believed to be the cancer that occurs most frequently in women worldwide, taking the lives of it’s the victims. In early diagnosis aids the patients to survive under greater probability. Several existing studies utilize diagnostic mechanisms via histopathology image for early identification of breast tumors. However, it increases the medical costs and consumes the time. Thus, in order to accurately classify the breast tumor, this study suggests a novel explainable DL technique. Using this technique, better accuracy is achieved while performing classifications. Improved accuracy may greatly help the medical practitioners for classifying breast cancer effectively. Initially, adaptive unsharp mask filtering (AUMF) technique is proposed to remove the noise and enhance the quality of the image. Finally, Explainable Soft Attentive EfficientNet (ESAE-Net) technique is introduced to classify the breast tumor (BT). Four explainable algorithms are investigated for improved visualizations over the BTs: Gradient-Weighted Class Activation Mapping (Grad-CAM) Shapley additive explanations (SHAP), Contextual Importance and Utility (CIU), and Local Interpretable Model-Agnostic Explanations (LIME). The suggested approach uses two publicly accessible images of breast histopathology and is carried out on a Python platform. Performance metrics such as time complexity, False Discovery Rate (FDR), accuracy, and Mathew’s correlation coefficient (MCC) are examined and contrasted with traditional research. In the experimental section, the proposed obtains an accuracy of 97.85% for dataset 1 and accuracy of 98.05% for dataset 2. In comparison with other existing methods, the proposed method is more efficient while using ESAE-Net for classifying the Breast cancer.

Toward explainable AI (XAI) for mental health detection based on language behavior.

Advances in artificial intelligence (AI) in general and Natural Language Processing (NLP) in particular are paving the new way forward for the automated detection and prediction of mental health disorders among the population. Recent research in this area has prioritized predictive accuracy over model interpretability by relying on deep learning methods. However, prioritizing predictive accuracy over model interpretability can result in a lack of transparency in the decision-making process, which is critical in sensitive applications such as healthcare. There is thus a growing need for explainable AI (XAI) approaches to psychiatric diagnosis and prediction. The main aim of this work is to address a gap by conducting a systematic investigation of XAI approaches in the realm of automatic detection of mental disorders from language behavior leveraging textual data from social media. In pursuit of this aim, we perform extensive experiments to evaluate the balance between accuracy and interpretability across predictive mental health models. More specifically, we build BiLSTM models trained on a comprehensive set of human-interpretable features, encompassing syntactic complexity, lexical sophistication, readability, cohesion, stylistics, as well as topics and sentiment/emotions derived from lexicon-based dictionaries to capture multiple dimensions of language production. We conduct extensive feature ablation experiments to determine the most informative feature groups associated with specific mental health conditions. We juxtapose the performance of these models against a "black-box" domain-specific pretrained transformer adapted for mental health applications. To enhance the interpretability of the transformers models, we utilize a multi-task fusion learning framework infusing information from two relevant domains (emotion and personality traits). Moreover, we employ two distinct explanation techniques: the local interpretable model-agnostic explanations (LIME) method and a model-specific self-explaining method (AGRAD). These methods allow us to discern the specific categories of words that the information-infused models rely on when generating predictions. Our proposed approaches are evaluated on two public English benchmark datasets, subsuming five mental health conditions (attention-deficit/hyperactivity disorder, anxiety, bipolar disorder, depression and psychological stress).

A Multivariate Time Series Analysis of Electrical Load Forecasting Based on a Hybrid Feature Selection Approach and Explainable Deep Learning

In the smart grid paradigm, precise electrical load forecasting (ELF) offers significant advantages for enhancing grid reliability and informing energy planning decisions. Specifically, mid-term ELF is a key priority for power system planning and operation. Although statistical methods were primarily used because ELF is a time series problem, deep learning (DL)-based forecasting approaches are more commonly employed and successful in achieving precise predictions. However, these DL-based techniques, known as black box models, lack interpretability. When interpreting the DL model, employing explainable artificial intelligence (XAI) yields significant advantages by extracting meaningful information from the DL model outputs and the causal relationships among various factors. On the contrary, precise load forecasting necessitates employing feature engineering to identify pertinent input features and determine optimal time lags. This research study strives to accomplish a mid-term forecast of ELF study load utilizing aggregated electrical load consumption data, while considering the aforementioned critical aspects. A hybrid framework for feature selection and extraction is proposed for electric load forecasting. Technical term abbreviations are explained upon first use. The feature selection phase employs a combination of filter, Pearson correlation (PC), embedded random forest regressor (RFR) and decision tree regressor (DTR) methods to determine the correlation and significance of each feature. In the feature extraction phase, we utilized a wrapper-based technique called recursive feature elimination cross-validation (RFECV) to eliminate redundant features. Multi-step-ahead time series forecasting is conducted utilizing three distinct long-short term memory (LSTM) models: basic LSTM, bi-directional LSTM (Bi-LSTM) and attention-based LSTM models to accurately predict electrical load consumption thirty days in advance. Through numerous studies, a reduction in forecasting errors of nearly 50% has been attained. Additionally, the local interpretable model-agnostic explanations (LIME) methodology, which is an explainable artificial intelligence (XAI) technique, is utilized for explaining the mid-term ELF model. As far as the authors are aware, XAI has not yet been implemented in mid-term aggregated energy forecasting studies utilizing the ELF method. Quantitative and detailed evaluations have been conducted, with the experimental results indicating that this comprehensive approach is entirely successful in forecasting multivariate mid-term loads.