Annually, web search engine providers spend a lot of money on re-ranking documents in search engine result pages (SERP). Click models provide advantageous information for re-ranking documents in SERPs through modeling interactions among users and search engines. Here, three modules are employed to predict users' clicks on SERPs simultaneously, the first module tries to predict users' click behaviors using Probabilistic Graphical models, the second module is a Time-series Deep Neural Click model which predicts users' clicks on documents and finally, the third module is a similarity-based measure which creates a graph of document-query relations and uses SimRank Algorithm to predict the similarity. After running these three simultaneous processes, three click probability values are fed to an MLP classifier as inputs. The MLP classifier learns to decide on top of the three preceding modules, then it predicts a probability value which shows how probable a document is to be clicked by a user. The proposed system is evaluated on the Yandex dataset as a standard click log dataset. The results demonstrate the superiority of our model over the well-known click models in terms of perplexity.

Since a robust and accurate classification of tumors is essential for successful treatment of cancer, classification of DNA microarray data has been widely used in effective diagnosis of cancers and some other biological maladies. This paper presents a new Ensemble machine learning model for development of robust microarray data classification. The main idea of the proposed model is to build a homogeneous Ensemble of base classifiers using different gene subsets to predict the class of each sample. We used stacked generalization as the combination scheme to learn the base classifiers' behavior for each sample. The performance of the proposed model has been evaluated over four publicly available cancer microarray datasets. Experimental results have demonstrated that not only the suggested classifier greatly outperforms existing conventional machine learning methods, but also it is notably more accurate and robust than some traditional Ensemble based classifiers and its performance could be compared to some recently introduced outstanding Ensemble-based methods.

Steganalysis is an interesting classi cation problem to discriminate the images, including hidden messages from the clean ones. There are many methods, including deep CNN networks, to extract  ne features for this classi cation task. Also, some researches have been conducted to improve the  nal classi er. Some state-of-the-art methods use Ensemble of networks by a voting strategy to achieve more stable performance. In this paper, a selection phase is proposed to  lter improper networks before any voting. This  ltering is done by a binary relevance multi-label classi cation approach. Xu-Net and ResT-Net, the most famous state-of-the-art Steganalysis Ensemble models, are considered as the base networks for feature extraction. The Logistic Regression (LR) is chosen here as the last layer of the networks for classi cation. One large-margin Fisher's linear discriminant (FLD) classi er is trained for each one of the networks to measure its suitability in classifying the query image. The proposed method with di erent approaches is applied on the BOSSbase dataset and compared to traditional voting and some state-of-the-art related Ensemble techniques. The results show signi cant accuracy improvement of the proposed method in comparison with others.

Background: Human leukocyte antigens (HLAs) play a pivotal role in orchestrating the host’s immune response, offering a promising avenue with reduced adverse effects compared to conventional treatments. Cancer immunotherapies use HLA class I molecules for T cells to recognize tumor antigens, emphasizing the importance of identifying peptides that bind effectively to HLAs. Computer modeling of HLA-peptide binding speeds up the search for immunogenic epitopes, which enhances the prospect of personalized medicine and targeted therapies. The Immune Epitope Database (IEDB) is a vital repository, housing curated immune epitope data and prediction tools for HLA-peptide binding. It can be challenging for immunologists to choose the best tool from the IEDB for predicting HLA-peptide binding. This has led to the creation of consensus-based methods that combine the results of several predictors. One of the major challenges in these methods is how to effectively integrate the results from multiple predictors.Objectives: Previous consensus-based methods integrate at most three tools by relying on simple strategies, such as selecting prediction methods based on their proximity to HLA in training data. In this study, we introduce HLAPepBinder, a novel consensus approach using Ensemble machine learning methods to predict HLA-peptide binding, addressing the challenges biologists face in model selection. Materials and Methods: The key contribution is the development of an automatic pipeline named HLAPepBinder that integrates the predictions of multiple models using a random forest approach. Unlike previous approaches, HLAPepBinder seamlessly integrates results from all nine predictors, providing a comprehensive and accurate predictive framework. By combining the strengths of these models, HLAPepBinder eliminates the need for manual model selection, providing a streamlined and reliable solution for biologists. Results: HLAPepBinder offers a practical and high-performing alternative for HLA-peptide binding predictions, outperforming both traditional methods and complex deep learning models. Compared to the recently introduced transformer-based model, TranspHLA, which requires substantial computational resources, HLAPepBinder demonstrates superior performance in both prediction accuracy and resource efficiency. Notably, it operates effectively in limited computational environments, making it accessible to researchers with minimal resources. The codes are available online at https://github.com/CBRC-lab/HLAPepBinder.Conclusion: Our study introduces a novel Ensemble-learning model designed to enhance the accuracy and efficiency of HLA-peptide binding predictions. Due to the lack of reliable negative data and the typical assumption of unknown interactions being negative, we focus on analyzing the unknown HLA-peptide bindings in the test set that our model predicts with 100% certainty as positive bindings. Using HLAPepBinder, we identify 26 HLA-peptide pairs with absolute prediction confidence. These predictions are validated through a multi-step pipeline involving literature review, BLAST sequence similarity analysis, and molecular docking studies. This comprehensive validation process highlights HLAPepBinder’s ability to make accurate and reliable predictions, contributing significantly to advancements in immunotherapy and vaccine development.

The bankruptcy of corporations causes huge losses for investors, managers, creditors, employees, suppliers, and customers. If someone understands the reason for the corporate's bankruptcy, then he can save the corporate from certain death with the necessary planning. Therefore, bankruptcy forecasting is the most important prerequisite for bankruptcy prevention. Due to this issue, the main aim of this article is the prediction of the economic bankrupt-cy of corporations in the Tehran Stock Exchange using group machine learn-ing algorithms. Financial ratios have been used as independent variables and healthy and bankrupt corporations as research dependent variables. The statistical population of the study is the information of financial statements of corporations on the Tehran Stock Exchange from the years 2004 to 2021. In this study, sampling is not used and corporations include two groups healthy and bankrupt. The bankrupt and non-bankrupt groups are selected based on the threshold of the Springate model. The research findings indicate that the accuracy of predicting the bankruptcy of corporations in the group learning model by stacking method is higher than other used models where the AUC and Accuracy Ratio were 0.9276 and 0.8247, respectively.

In recent years, the growth of social networks and, consequently, the increasing content of these networks have led people to use others’ opinions to make decisions for the purchase and use of products, services or even political choices. Given the fact that users' comments are textual and their reading and summarizing is timely and difficult, the automation of the extraction of opinions and sentiments of users' comments is one of the suggested solutions for online sales sites to provide more efficient services to customers for better decision making. Sentiment analysis or opinion mining is a process where people's opinions, feelings and attitudes are extracted in relation to a particular subject and are recognized as a branch of the text mining. The results of sentiment analysis can be used in recommender systems to provide more effective shopping suggestions. Information derived from the opinion mining can be used in a variety of fields such as libraries for better choices and purchases based on the users' real opinions. In this research, a system for automatically categorizing the sentiments expressed in the opinions of the buyers of the Amazon book website is presented. The system is designed using Ensemble voting models to analyze the sentiment of Amazon users' comments. For all analyses, Python text mining packages are used. In Ensemble method two methods are used: majority voting and weight-based voting. In the weighting method, a greater weight is assigned to a classifier by higher accuracy. By comparing the performance of the results, the weighting model is chosen as the final model for making the sentiment analysis. Results show that the proposed system can automatically classify positive and negative comments with an accuracy of over 80%.

Climate change has significantly increased the frequency and intensity of heat stress and has more effects than increasing average temperature. This study has investigated the spatial distribution of the universal thermal climate index (UTCI) during historical and future periods in Iran. The UTCI (°C) refers to “the isothermal air temperature of the reference condition that would elicit the same dynamic response (strain) of the physiological model” (Jendritzky et al., 2012). In this way, the UTCI is an equivalent temperature, similar to PT. The thermal impact of the meteorological conditions is compared to the one of a standardized reference “indoor” environment with RH = 50% (Ta < 29 °C), WS = 0.5 m s−1, pa = 20 hPa (Ta < 29 °C), and Tmrt = Ta (Shin et al. 2022). Three variables of daily temperature, relative humidity, and wind speed from two sets of data, including 124 meteorological stations and five models from the Coupled model Intercomparison Project phase 6 (CMIP6) model, including GFDL-ESM4, IPSL-CM6A-LR, MPI-ESM1-2-HR, MRI-ESM2-0, and UKESM1-0-LL were investigated with a horizontal resolution of 0.5o. Then, an Ensemble model (CMIP6-MME) was generated from these five models using the independent weighted mean (IWM) method. The performances of individual models and the generated Ensemble model were examined by Taylor's diagram. The results showed that the multi-model Ensemble has higher performance than individual models for all three variables. The results revealed that the spatial distribution of the seasonal averages of the UTCI index has significant variability in Iran, and the variability of this index is affected by the latitude, complex topography, and distance to water resources in Iran. In general, heat stress will increase significantly in Iran by the end of the century. So, we will witness a significant decrease in areas with no heat stress until the end of this century. On the contrary, strong to very strong heat stress events will increase significantly in the country at the end of the century. While the areas with no thermal stress show a spatial displacement to mountainous regions and higher latitudes. These results show that effective adaptation methods should be taken to adapt to global warming and reduce its consequences to avoid the adverse effect of increasing heat stress events in Iran. The results show the overall increasing trend of Iran's heat stress in the near and far future. The highest increase in heat stress anomalies (13.3 degrees Celsius in winter during the far future period under the SSP5-8.5 scenario) can be found in the northwest and west of the country. The increasing intensity of heat stress in the western and northwestern parts of Iran may be related to elevation-dependent warming (EDW).