Selected Performance Measures Research Articles

Association of plasma proteomics with incident coronary heart disease in individuals with and without type 2 diabetes: results from the population-based KORA study.

Coronary heart disease (CHD) is a major global health concern, especially among individuals with type 2 diabetes (T2D). Given the crucial role of proteins in various biological processes, this study aimed to elucidate the aetiological role and predictive performance of protein biomarkers on incident CHD in individuals with and without T2D. The discovery cohort included 1492 participants from the Cooperative Health Research in the Region of Augsburg (KORA) S4 study with 147 incident CHD cases (45 vs. 102 cases in the group with T2D and without T2D, respectively) during 15.6 years of follow-up. The validation cohort included 888 participants from the KORA-Age1 study with 70 incident CHD cases (19 vs. 51 cases in the group with T2D and without T2D, respectively) during 6.9 years of follow-up. We measured 233 plasma proteins related to cardiovascular disease and inflammation using proximity extension assay technology. Associations of proteins with incident CHD were assessed using Cox regression and Mendelian randomization (MR) analysis. Predictive models were developed using priority-Lasso and were evaluated on top of Framingham risk score variables using the C-index, category-free net reclassification index (cfNRI), and relative integrated discrimination improvement (IDI). We identified two proteins associated with incident CHD in individuals with and 29 in those without baseline T2D, respectively. Six of these proteins are novel candidates for incident CHD. MR suggested a potential causal role for hepatocyte growth factor in CHD development. The developed four-protein-enriched model for individuals with baseline T2D (ΔC-index: 0.017; cfNRI: 0.253; IDI: 0.051) and the 12-protein-enriched model for individuals without baseline T2D (ΔC-index: 0.054; cfNRI: 0.462; IDI: 0.024) consistently improved CHD prediction in the discovery cohort, while in the validation cohort, significant improvements were only observed for selected performance measures (with T2D: cfNRI: 0.633; without T2D: ΔC-index: 0.038; cfNRI: 0.465). This study identified novel protein biomarkers associated with incident CHD in individuals with and without T2D and reaffirmed previously reported protein candidates. These findings enhance our understanding of CHD pathophysiology and provide potential targets for prevention and treatment.

An ANN Model to Predict Select Performance Measures for Job Scheduling System

The purpose of this work is to design and develop an artificial neural network (ANN) model for prediction of the select performance measures (PMs) of a single machine job scheduling (SMJS) system, under the influence of varying levels of input variables. Ten thousand scheduling problems were randomly generated and for each problem values of five PMs were computed which were used to train, test, and validate different ANN models in order to develop the optimal ANN model for accurate prediction of the PMs for a given set of input variables. An ANN model with two hidden layers having 16 neurons in each hidden layer was found to be the optimum model for prediction of the PMs as it resulted in the minimum mean squared error. The actual and predicated values of the PMs obtained from the optimum ANN model were compared and it was found that they were in close agreement. The ANN model developed in this study may be used by the managers, practitioners and other decision makers to define appropriate values of input variables that will yield better PMs. Further, the approach used in the study to develop ANN model may be used for development of similar ANN models for other manufacturing systems. This study provides an ANN-based prediction model to predict PMs of the single machine job scheduling system which may however, be modified as per the requirements of other manufacturing systems to predict their PMs.

DeepNet model empowered cuckoo search algorithm for the effective identification of lung cancer nodules.

Globally, lung cancer is a highly harmful type of cancer. An efficient diagnosis system can enable pathologists to recognize the type and nature of lung nodules and the mode of therapy to increase the patient's chance of survival. Hence, implementing an automatic and reliable system to segment lung nodules from a computed tomography (CT) image is useful in the medical industry. This study develops a novel fully convolutional deep neural network (hereafter called DeepNet) model for segmenting lung nodules from CT scans. This model includes an encoder/decoder network that achieves pixel-wise image segmentation. The encoder network exploits a Visual Geometry Group (VGG-19) model as a base architecture, while the decoder network exploits 16 upsampling and deconvolution modules. The encoder used in this model has a very flexible structural design that can be modified and trained for any resolution based on the size of input scans. The decoder network upsamples and maps the low-resolution attributes of the encoder. Thus, there is a considerable drop in the number of variables used for the learning process as the network recycles the pooling indices of the encoder for segmentation. The Thresholding method and the cuckoo search algorithm determines the most useful features when categorizing cancer nodules. The effectiveness of the intended DeepNet model is cautiously assessed on the real-world database known as The Cancer Imaging Archive (TCIA) dataset and its effectiveness is demonstrated by comparing its representation with some other modern segmentation models in terms of selected performance measures. The empirical analysis reveals that DeepNet significantly outperforms other prevalent segmentation algorithms with 0.962 ± 0.023% of volume error, 0.968 ± 0.011 of dice similarity coefficient, 0.856 ± 0.011 of Jaccard similarity index, and 0.045 ± 0.005s average processing time.

Clinical Performance Measures for Emergency Department Care for Adults With Intracranial Hemorrhage

Though select inpatient-based performance measures exist for the care of patients with nontraumatic intracranial hemorrhage, emergency departments lack measurement instruments designed to support and improve care processes in the hyperacute phase. To address this, we propose a set of measures applying a syndromic (rather than diagnosis-based) approach informed by performance data from a national sample of community EDs participating in the Emergency Quality Network Stroke Initiative. To develop the measure set, we convened a workgroup of experts in acute neurologic emergencies. The group considered the appropriate use case for each proposed measure: internal quality improvement, benchmarking, or accountability, and examined data from Emergency Quality Network Stroke Initiative-participating EDs to consider the validity and feasibility of proposed measures for quality measurement and improvement applications. The initially conceived set included 14 measure concepts, of which 7 were selected for inclusion in the measure set after a review of data and further deliberation. Proposed measures include 2 for quality improvement, benchmarking, and accountability (Last 2 Recorded Systolic Blood Pressure Measurements Under 150 and Platelet Avoidance), 3 for quality improvement and benchmarking (Proportion of Patients on Oral Anticoagulants Receiving Hemostatic Medications, Median ED Length of Stay for admitted patients, and Median Length of Stay for transferred patients), and 2 for quality improvement only (Severity Assessment in the ED and Computed Tomography Angiography Performance). The proposed measure set warrants further development and validation to support broader implementation and advance national health care quality goals. Ultimately, applying these measures may help identify opportunities for improvement and focus quality improvement resources on evidence-based targets.

NIGERIA'S BANKING SECTOR REFORMS AND COMMERCIAL BANKS' PERFORMANCE: A SIMPLE TEST OF THE MODIFIED PANTULA PRINCIPLE

This study examined the achievement of organizational goals by deposit money banks following the period of banking sector reform. Annual time series data from the banking industry and the Nigerian economy for the period 1990 to 2020 were used to implement a simplified version of the modified Pantula Principle (an improved cointegrating technique). Three sets of reforms and bank performance measures, including bank-specific, industry-specific, and macroeconomic indicators, were identified and adopted. The empirical findings obtained from our analysis revealed that there are no long-run connections between selected performance measures and some bank-specific, industry-specific, and macroeconomic variables, implying that the reforms had no long-term impact on the statutory achievement of the banks' goals in Nigeria during and beyond the era.
 Keywords: Pantula Principle; Banking Sector Reforms; Banks' Performance; Nigeria.
 JEL Classification : C18,G21,G28

Artificial intelligence in BreastScreen Norway: a retrospective analysis of a cancer-enriched sample including 1254 breast cancer cases

ObjectivesTo compare results of selected performance measures in mammographic screening for an artificial intelligence (AI) system versus independent double reading by radiologists.MethodsIn this retrospective study, we analyzed data from 949 screen-detected breast cancers, 305 interval cancers, and 13,646 negative examinations performed in BreastScreen Norway during the period from 2010 to 2018. An AI system scored the examinations from 1 to 10, based on the risk of malignancy. Results from the AI system were compared to screening results after independent double reading. AI score 10 was set as the threshold. The results were stratified by mammographic density.ResultsA total of 92.7% of the screen-detected and 40.0% of the interval cancers had an AI score of 10. Among women with a negative screening outcome, 9.1% had an AI score of 10. For women with the highest breast density, the AI system scored 100% of the screen-detected cancers and 48.6% of the interval cancers with an AI score of 10, which resulted in a sensitivity of 80.9% for women with the highest breast density for the AI system, compared to 62.8% for independent double reading. For women with screen-detected cancers who had prior mammograms available, 41.9% had an AI score of 10 at the prior screening round.ConclusionsThe high proportion of cancers with an AI score of 10 indicates a promising performance of the AI system, particularly for women with dense breasts. Results on prior mammograms with AI score 10 illustrate the potential for earlier detection of breast cancers by using AI in screen-reading.Key Points• The AI system scored 93% of the screen-detected cancers and 40% of the interval cancers with AI score 10.• The AI system scored all screen-detected cancers and almost 50% of interval cancers among women with the highest breast density with AI score 10.• About 40% of the screen-detected cancers had an AI score of 10 on the prior mammograms, indicating a potential for earlier detection by using AI in screen-reading.

Abstract WP49: Clinical Performance Measures For Emergency Department Care For Adults With Nontraumatic Intracranial Hemorrhage

Study Objective: Though select inpatient-based performance measures exist for the care of patients with nontraumatic intracranial hemorrhage (ICH), Emergency Departments (EDs) lack measurement instruments designed to support and improve care processes in the hyperacute phase. To address this, we propose a set of measures applying a syndromic (rather than diagnosis-based) approach informed by performance data from a national sample of community EDs participating in the Emergency Quality Network Stroke Initiative (E-QUAL). Methods: We convened a workgroup of experts in acute neurologic emergencies. The group considered the appropriate use case for each proposed measure (internal quality improvement [QI], benchmarking, or accountability) and examined data from E-QUAL-participating EDs to consider the validity and feasibility of proposed measures. The initially conceived set included 14 measure concepts of which 7 were selected for inclusion in the measure set after review of data and further deliberation. To inform the process, we collected data from E-QUAL participating EDs. EDs used a web-based portal to submit data from chart reviews on patients with ICH during 2020-2021. Descriptive statistics characterize performance and identify variation. Results: Proposed measures include 2 for QI, benchmarking, and accountability, 3 for QI and benchmarking and 2 for QI only (Table). Of the E-QUAL participating EDs, 35 had sufficient case volume (≥25) and were included for review of preliminary performance data. Median annual ED volume was 44,000 (IQR 31,337-67,000) and 16 (48%) reported typically transferring patients with ICH. The Table includes available performance data. Conclusion: Application of these measures may identify opportunities for improvement and focus QI resources on evidence-based targets. The proposed measures warrant further development and validation to support broader implementation and advance national healthcare quality goals.

The impact of the pharmaceutical industry on the innovation performance of European countries

There are significant differences in innovation performance between countries. Additionally, the pharmaceutical sector is stronger in some countries than in others. This suggests that the development of the pharmaceutical industry can influence a country's innovation performance. Using the Global Innovation Index and selected performance measures of the pharmaceutical sector, this study examines how the pharmaceutical sector influences the innovation performance of countries from the European context. The dataset of 27 European countries was analysed using simple, and multiple linear regressions and Pearson correlation. Our findings show that only three indicators of the pharmaceutical industry, more precisely pharmaceutical Research and Development, pharmaceutical exports, and pharmaceutical employment explain the innovation performance of a country largely. Pharmaceutical Research and Development and exports have a significant positive impact on a country's innovation performance, whereas employment in the pharmaceutical industry has a slightly negative impact. Additionally, global innovation performance has been found to positively influence life expectancy. We further outline the implications and possible policy directions based on these findings.

Interpretable meta-score for model performance

Benchmarks are an integral part of machine learning development. However, the most common benchmarks share several limitations. For example, the difference in performance between two models has no probabilistic interpretation, it makes no sense to compare such differences between data sets and there is no reference point that indicates a significant performance improvement. Here we introduce an Elo-based predictive power meta-score that is built on other performance measures and allows for interpretable comparisons of models. Differences between this score have a probabilistic interpretation and can be compared directly between data sets. Furthermore, this meta-score allows for an assessment of ranking fitness. We prove the properties of the Elo-based predictive power meta-score and support them with empirical results on a large-scale benchmark of 30 classification data sets. Additionally, we propose a unified benchmark ontology that provides a uniform description of benchmarks. The performance of machine learning models is usually compared via the mean value of a selected performance measure such as the area under the receiver operating characteristic curve on a specific benchmark data set. However, this measure, its mean value or even relative differences between models do not provide a good prediction of whether the results can translate to other data sets. Gosiewska and colleagues present here a comparison based on Elo ranking, which offers a probabilistic interpretation of how much better one model is than another.

Defining performance thresholds for effective management of biodiversity within protected areas.

Performance thresholds are an important tool for determining successful conservation outcomes. They provide an objective means of defining good ecological condition and have been endorsed as an essential part of best practice in protected area (PA) management within the International Union for the Conservation of Nature Green List of Protected and Conserved Areas Standard. With a growing number of PAs attaining Green List status globally, thresholds developed by PAs on the Green List present an excellent resource with which to identify the attributes of well-defined performance thresholds. We examined 349 thresholds associated with PAs on the Green List to determine whether they were specific and measurable (i.e., factors recognized as essential for setting well-defined targets). We assessed whether thresholds were defined quantitatively and whether definitions included ambiguous terms (e.g., stable numbers). We identified six different ways thresholds were expressed and found that many thresholds were expressed as management objectives, rather than ecological condition thresholds, although this trend improved over time. Approximately one-half of the performance thresholds lacked the necessary specificity to delineate successful outcomes. Our results enabled us to develop a checklist of information required to set robust performance thresholds. Recommendations include that thresholds should be quantitatively defined, including quantitative estimates of the limits of acceptable change (LAC) around the target condition. To ensure transparency, a rationale and associated evidence should be provided to support the threshold and the LAC. When accompanied by a rationale and quantitative estimate of the current condition of the value, unambiguously defined thresholds with a quantitative LAC provide an objective means of demonstrating that successful conservation outcomes have been achieved. These recommendations will help conservation managers apply the Green List Standard and improve the measurement of conservation outcomes more broadly.

Long-term forecasting of hourly retail customer flow on intermittent time series with multiple seasonality

In this study, we address a demanding time series forecasting problem that deals simultaneously with the following: (1) intermittent time series, (2) multi-step ahead forecasting, (3) time series with multiple seasonal periods, and (4) performance measures for model selection across multiple time series. Current literature deals with these types of problems separately, and no study has dealt with all these characteristics simultaneously. To fill this knowledge gap, we begin by reviewing all the necessary existing literature relevant to this case study with the goal of proposing a framework capable of achieving adequate forecast accuracy for such a complex problem. Several adaptions and innovations have been conducted, which are marked as contributions to the literature. Specifically, we proposed a weighted average forecast combination of many cutting-edge models based on their out-of-sample performance. To gather strong evidence that our ensemble model works in practice, we undertook a large-scale study across 98 time series, rigorously assessed with unbiased performance measures, where a week seasonal naïve was set as a benchmark. The results demonstrate that the proposed ensemble model achieves eye-catching forecasting accuracy.

FINANCIAL PERFORMANCE – DETERMINANTS AND INTERDEPENDENCIES BETWEEN MEASUREMENT INDICATORS

Purpose – the study has a dual purpose. First, to assess the impact of the most important determinants of financial performance, which have been measured through four generations of indicators. In addition, the study provides the first quantification of interdependencies between different financial performance measures: profit margin (PM), profit growth rate (PGR), return on assets (ROA), return on equity (ROE), and economic value added (EVA). Research methodology – the primary data was collected from the AMADEUS database. Empirical research was conducted on a relatively homogeneous sample from the automotive industry, using the panel data method for the period 2010–2019. Two models were tested. The first model highlights the relationships between performance measures and selected determinants. The second model highlights the relationship between the different performance measures and the determinants used in the first model. Findings – the determinants analysed have different influences on the selected performance measures. For example, in the first model, the results statistically significant indicated the following. The current ratio has a positive influence on ROA, but a negative one on ROE and EVA. Gearing has a negative influence on PM and ROA, but a positive one on EVA. The growth rate of sales has a positive influence on PM, but a negative one on ROA and EVA. The size of the company has a positive influence on three performance measures (PM, ROA, and EVA). Regarding the relationships between the different performance measures (second model), the research indicates that EVA is negatively influenced by PGR and ROA. In this model, the determinants analysed maintain their meaning and intensity of influences. Research limitations – the article has several limitations. The representativeness of the results is valuable only at the level of the researched industry. In addition, it should be noted that the analyses are focused only on financial performance, assessed by accounting measures. The authors are considering conducting comparative analyses at the level of fields/branches of activity to capture not only the impact of determinants on financial performance but also to assess organizational resilience. Practical implications – The research provides clues to managers and financial decision-makers to increase the financial performance of the companies they lead. Originality/value – the originality of the study lies in the presented methodological approach. Unlike previous research, which usually evaluated performance on only one indicator, this paper aims to assess the impact of the most important determinants on five performance measures. In addition, the analysis of the interdependencies between the different performance measures is another novelty of this research.

Development and Structural Organization of Mexico’s Mangrove Monitoring System (SMMM) as a Foundation for Conservation and Restoration Initiatives: A Hierarchical Approach

Mangroves provide ecosystem services worth billions of dollars worldwide. Although countries with extensive mangrove areas implemented management and conservation programs since the 1980s, the global area is still decreasing. To recuperate this lost area, both restoration and rehabilitation (R/R) projects have been implemented but with limited success, especially at spatial scales needed to restore functional properties. Monitoring mangroves at different spatial scales in the long term (decades) is critical to detect potential threats and select cost-effective management criteria and performance measures to improve R/R program success. Here, we analyze the origin, development, implementation, and outcomes of a country-level mangrove monitoring system in the Neotropics covering >9000 km2 over 15 years. The Mexico’s Mangrove Monitoring System (SMMM) considers a spatiotemporal hierarchical approach as a conceptual framework where remote sensing is a key component. We analyze the role of the SMMM’s remote sensing products as a “hub” of multi- and interdisciplinary ecological and social-ecological studies to develop national priorities and inform local and regional mangrove management decisions. We propose that the SMMM products, outcomes, and lessons learned can be used as a blueprint in other developing countries where cost-effective R/R projects are planned as part of mangrove protection, conservation, and management programs.

Psychophysiological Adaptation of International Students to Learning Abroad

The article describes the problems international students encounter during psychophysiological adaptation. As the modern higher education system is characterized by academic mobility, and international students from countries with quite different geoclimatic and cultural traditions have to undergo an adaptation period, the problem under study is topical. The study was conducted at Kharkiv National Medical University (Ukraine) in 2020/2021. The research involved 1st, 2nd and 3rd-year international students (from 50 countries). The peculiarities of international students’ psychophysiological states were assessed during the stages of initial adaptation, stable adaptation and final adaptation. This study aims to provide a comprehensive analysis of the factors of the natural and socio-pedagogical environment that affect the psychological adaptation process of international students. The present study provides new information on the acclimatization patterns of international students who study in Ukraine, as well as its functional significance on selected performance measures. General theoretical methods, research methods and methods of statistical data processing have been employed. The results reveal that the destabilizing factors that negatively affect international students’ adaptation process include frequent hypothermia, dietary changes, inadequate and irregular food intake, insufficient physical activity and excessive mental stress. Practical recommendations to decrease these negative factors are proposed.

A Quality Improvement Initiative to Reduce Postoperative Delirium among Cardiac Surgery Patients.

Patients following cardiac surgery commonly experience post-operative delirium (POD) during their postoperative hospital stay. A multifaceted, specialty wide, quality improvement (QI) project was undertaken for patients experiencing POD. The goal was to develop a reduction in POD care bundle (rPOD-a structured patient care program) that encompasses efficient preoperative risk factor identification and a postoperative patient-care process to ensure early POD identification and treatment. The following steps were taken to implement the rPOD care bundle including: (a) Developing a quality driven, evidence-based guideline for the perioperative cardiac surgery health care team, (b) identifying and addressing local barriers to implementation, (c) selecting performance measures to assess intervention adherence and patient outcomes, and (d) ensuring that all patients receive the interventions through staff engagement and education, and regular project evaluation. Trends of process measures and quality improvement measures were examined. An increasing trend in the rate of postoperative delirium screening during implementation of rPOD intervention was demonstrated. This quality improvement study provides a bases for future postoperative delirium reduction interventions.

Sustainability Performance of Port Infrastructure—A Case Study of a Quay Wall

Ports form a vital link in the global maritime supply chain, and adherence to the UN sustainable goals in each port development and operation is all-important. Improving the sustainability performance of port infrastructure requires identifying all relevant aspects of sustainability, defining suitable performance measures, applying tools for quantification, and proposing intervention measures if needed. We present a framework for assessing and managing the sustainability performance of port infrastructure incorporating these elements. We apply the framework to quay walls in the Port of Rotterdam (PoR); however, the information to assess only four out of fourteen sustainability themes was available. This fact underlines the need for sustainability monitoring and reporting. We also present a case study of a quay wall in PoR. Based on the vision of PoR and literature, the sustainability theme ‘air pollutants’ was given priority, the selected performance measure being the Greenhouse Gas (GHG) emissions. Dubocalc was chosen as a quantification tool and estimated a total emission of 1.9 kt of CO2eq for a 100-m length of quay wall over a 100-year lifecycle. The measures proposed to achieve climate neutrality in 2050 include using renewable energy for the Impressed Current Cathodic Protection and the temporary drainage systems, electrified transport and machinery, hydrogen as dredging fuel, and steel with hydrogen as a reduction-agent in the future. These could result in a total GHG reduction of 86% by the year 2050.

Default on a credit prediction using decision tree and ensemble learning techniques

Due to the high importance of credit risk management and the prediction of the default on a credit loan in banks in recent years in most developed countries. This research experimented with three machine learning algorithms which are decision tree and ensemble techniques based on a decision tree-like random forest which is bootstrap aggregation machine learning model and adaptive boosting is a boosting machine learning model. All models were used to perform a binary classification to classify the clients into trusted clients who would pay back the debts who are creditable and a none trusted clients who won’t pay the debts In the time which are who are not creditable. All the models gave nearly the same results based on the selected performance measures which are precision, recall, and f1 measure. There was no significant change when they performed on a real class imbalanced dataset of the default credit card clients of Taiwan.

APPLICATION POTENTIAL OF RESPONSE SURFACE METHOD ON ELECTRO DISCHARGE MACHINING OF AA6061–CENOSPHERE AMCs PREPARED BY COMPOCASTING METHOD

Cenosphere fly ash particles are incorporated into AA6061 alloys with different concentrations ranging from 0[Formula: see text]wt.% to 10[Formula: see text]wt.% using a modified semi-solid metal processing technique. X-ray diffraction patterns were recorded to analyze the morphology of the aluminum-based metal matrix composites (AMCs). The major diffraction peaks of Al, SiO2, Al2O3 and Fe2O3 are distinctly identified which revealed the presence of cenosphere particles and their integrity within the matrix is preserved. The high-resolution optical micrograph identifies the homogeneous distribution and uniform dispersion of the particles. Machinability of the prepared AMCs was investigated by electro discharge machining (EDM) using response surface methodology (RSM). Face-centered CCD of RSM was considered to design the number of experimental runs required. ANOVA was used to explore the influence of selected process parameters and their interactions on the performance characteristics of the systems by developing a second-order quadratic mathematical model for all the responses. Pulse on-time and pulse current were observed to be the most influencing independent variables of EDM system that affect the selected performance measures during spark erosion process. Finally, desirability function approach was employed to optimize the parameters. The optimal processing condition was identified as follows: pulse current: 6 A, pulse on-time: 1010[Formula: see text][Formula: see text]s, percentage of reinforcement: 2% and flushing pressure: 0.2 MPa. Very small percentages of deviation have been observed while comparing with the experimental results obtained for MRR (8.6%), TWR (10.3%) and SR (2.18%).

KPI Selection Using The AHP Method on SOE X

PT X (Persero), a state-owned enterprise under the Ministry of Indonesia SOEs (KBUMN), is required to set up the Key Performance Indicators (KPI) Regulations based on the Assessment Criteria for Performance Excellence (KPKU). The re-weighting of KPIs using the Analytical Hierarchy Process (AHP) method is conducted by selecting performance indicators based on five perspectives. Based on the study results, selected performance measures for financial and market perspectives are cash position, cash-to-cash time cycle, income, and financial operations efficiency. Selected performance measures for customer focus perspective are customer complaints, complaint management, effective complaint resolution, and warranty claims; customer satisfaction and dissatisfaction; and retention, acquisition, and loss of customers and customer accounts. Selected performance measures for product and process effectiveness perspective are product performance, work system performance, improved performance of administrative functions and other supporting functions; and indicators of internal responsiveness. Selected performance measures for the perspective of workforce focus are turnover and effectiveness of training, retraining, or cross-functional training. Selected performance measures for leadership, governance, and community responsibility perspectives are the company's contribution to social welfare, achievements worthy of note in this field; and environmental, legal, and regulatory compliance.

KP-Rank: a semantic-based unsupervised approach for keyphrase extraction from text data

Automatic key concept identification from text is the main challenging task in information extraction, information retrieval, digital libraries, ontology learning, and text analysis. The main difficulty lies in the issues with the text data itself, such as noise in text, diversity, scale of data, context dependency and word sense ambiguity. To cope with this challenge, numerous supervised and unsupervised approaches have been devised. The existing topical clustering-based approaches for keyphrase extraction are domain dependent and overlooks semantic similarity between candidate features while extracting the topical phrases. In this paper, a semantic based unsupervised approach (KP-Rank) is proposed for keyphrase extraction. In the proposed approach, we exploited Latent Semantic Analysis (LSA) and clustering techniques and a novel frequency-based algorithm for candidate ranking is introduced which considers locality-based sentence, paragraph and section frequencies. To evaluate the performance of the proposed method, three benchmark datasets (i.e. Inspec, 500N-KPCrowed and SemEval-2010) from different domains are used. The experimental results show that overall, the KP-Rank achieved significant improvements over the existing approaches on the selected performance measures.