Ranking of researchers based on their scientific impact in a scientific community is indeed a very crucial task. However, identifying the researchers ranking helps the scientific community in various decisions such as awarding scholarships, selection for tenure, awarding achievements, giving promotions, etc. In literature numerous parameters have been proposed for the ranking of researchers, such as publication count, citation count, coauthor count, h-index, and its extensions. The current state-of-the-art research delineates that no such universally accepted parameter exists which can identify the most influential researchers. Therefore, it is necessary to determine an optimal parameter that can effectively rank authors. Furthermore, to identify the best parameter, few of the researchers conducted evaluative surveys as reflected in the literature. In these evaluative surveys, the researchers utilized a limited number of indices on the small and imbalanced datasets, followed by fictional cases and scenarios, this has made it challenging to ascertain the relative importance and impact of each parameter in comparison to the others. This research evaluates the h-index and its thirty-two variants, which are based on the number of publications and citation count category used for ranking the authors. We have collected data from 1050 researchers working in the mathematical domain for our experimental purposes. For the benchmark dataset, we have collected the awardees’ data of the last two decades of four different societies belonging to mathematical domain. First and foremost, we have computed the correlations among the obtained values of the indices to assess their similarities and differences to evaluate indices. The result revealed that there is a high degree of correlation observed among h-index and it’s twenty-four different variants. However, some of the indices represented weak correlation, signifying that their rankings are highly dissimilar to those of other indices. Secondly, the position of awardees is checked in the top 10, top 50, and top 100 return records based on the ranking list of each index. The outcome of the last step divulges that A index, E index, H core citation, H2 lower index, K index, M index, and woginger index retrieved almost 80% of awardees in top 10% ranked list. Further, the analysis revealed that most of the winners (awards) were in the top tier, belonging to IMS, LMS, and AMS society returned by hg index, g index, k index, etc. indicating a relationship between the stated societies and the indices.

The assessment and evaluation of the academic influence of the researcher is a challenging task. This task allows the scientific community to make valuable decisions, such as identifying leading experts in a specific field, nominating candidates for scientific awards, awarding scholarships/grants, promoting researchers, and selecting tenure positions. Scientists have proposed various varied and multifaceted metrics to identify the most influential researchers. These metrics are the citation count, total number of publications, hybrid approaches, h-index, and variants of the h-index. Contemporary research in this domain shows that there is no universally accepted yardstick available for determining the finest parameter to recognize the most influential researcher within a particular domain. Moreover, to identify the potential metric, some researchers have conducted evaluation surveys. In these evaluation surveys, the researchers utilized a very limited number of indices on a small and imbalanced dataset as well as on fictional case scenarios, which makes it challenging to determine the significance and influence of each metric over the others. The present study computed fourteen distinct metrics based on the author-count. Our aim is to identify the potential metrics. For experimental purposes, we collected 1050 researchers’ data from the mathematics domain. For the benchmark dataset, we have collected the awardee’s data of the last three decades of four different societies of mathematics domain. To evaluate these metrics, we first computed the correlations among the obtained values of these metrics to assess their similarities and differences. The results showed a high degree of correlation among these metrics. However, some metrics represent weak correlations, signifying that their rankings are highly dissimilar to those of the other metrics. Furthermore, the position of award winners is checked in the top 10, 50, and 100 return records based on a ranked list of each metric. The potential value of each metric such as the hf metric, indicates that 60% of the awardees in the top 10% of the ranked list are associated with this, whereas the potential value of fractional g metric is linked with 49% of the awardees in the top 100% of the ranked list. Furthermore, it is further scrutinized that most of the award winners lie in a top position belonging to IMS, LMS, and AMS society return by fractional g-index, gf index, and gm index, which indicates that there is some relationship between these societies and metrics.

In the modern era, bibliometric parameters have become crucial components in evaluating academic productivity. In the last three decades, the scientific community has suggested a variety of bibliometric parameters to rank researchers, such as publication count, citation count, author count, h-index, and its variants. There is an ongoing debate within the scientific community regarding which index provides the most accurate ranking of authors. Currently, the state-of-the-art evaluation of these indices is based on hypothetical scenarios and the limited and unfair distribution of researcher records in the datasets. In addition, these indices are often evaluated using dissimilar datasets, making it challenging to accurately compare their contributions and significance. To comprehensively analyze the performance of each index, it should be evaluated on extensive and balanced datasets. This study focuses on evaluating the effectiveness of some h-index extensions, such as the platinum h, M quotient, AW, AR, V, Ha, Hc, and AWCR indices. For the experiments, this study employed a comprehensive and balanced dataset from the mathematics field to evaluate the performance of these indices. Analysis of these indices was conducted in three stages. First, the correlations among these indices were calculated. A weak correlation between indices suggests that the rankings of authors obtained from these indices are not the same, whereas a strong correlation suggests the opposite. Second, the analysis examines the occurrence of awardees in all ranked lists, using prestigious award winners from four mathematics societies as a benchmark. Based on the analysis, the study reveals that the Ar index performs the best among the evaluated indices, bringing the highest number of awardees to the top 10% of the ranked list, with a maximum of 80%. However, none of the indices can bring all the awardees to the top rankings, indicating that there is no single index that can provide a complete and accurate evaluation of the author’s ranking. Additionally, the study identified that the maximum awardees at the top of all ranked lists belong to the American Mathematical Society (AMS), suggesting that AMS may depend on these indices to determine its award recipients. Furthermore, the AWCR index is more suited to the AMS society by its contribution to returning the highest number of awardees belonging to the AMS society.

The measurement and assessment of academic performance is now a fact of scientific life. This assessment guides the scientific community in making significant judgments such as selecting appropriate candidates for various positions, nominating individuals for scientific awards, and awarding scholarships or grants. Several research assessment parameters have been proposed by researchers to identify the most influential scholars. In the literature, researchers have employed a combination of hypothetical and fictional scenarios, as well as manual approaches, to identify the best assessment parameters. Moreover, there is no established benchmark available for assessing these parameters. The current study employs an innovative machine learning approach, the Dynamic Random Forest with Brouta Optimizer, to prioritize the assessment metrics for researchers by calculating the importance score for each metric. Thirty different assessment metrics have been evaluated on a comprehensive dataset of researchers that contains awardees researchers and non-awardees researchers of three decades from (1990 to 2023). The main purpose of this evaluation is to determine the potential value and significance of each parameter relative to others. In addition, the position of awardees researchers is examined at different percentile ranges form Top 10% to Top 100% in the ranked lists of each metric. During the individual evaluation of each metric, we uncovered several intriguing patterns in the data. Our findings indicate that the normalized h-index is a particularly effective assessment metric for the impact evaluation of researchers in the domain of mathematics. An analysis has been conducted to explore the correlation between metrics and awarding societies, examining the associations between different metrics and specific awarding societies.

Recommendation systems are information filtering software that delivers suggestions about relevant stuff from a massive collection of data. Collaborative filtering approaches are the most popular in recommendations. The primary concern of any recommender system is to provide favorable recommendations based on the rating prediction of user preferences. In this article, we propose a novel discretization based framework for collaborative filtering to improve rating prediction. Our framework includes discretization-based preprocessing, chi-square based attribution selection, and K-Nearest Neighbors (KNN) based similarity computation. Rating prediction affords some basis for the judgment to decide whether recommendations are generated or not, subject to the ratio of performance of any recommendation system. Experiments on two datasets MovieLens and BookCrossing, demonstrate the effectiveness of our method.