Search results for: gaze metrics

Authors: Prajwal Prakash Vasisht, Sharath Rajamurthy, Nishanth Dara

Abstract:

In a supervised machine learning approach, metrics such as precision, accuracy, and coverage can be calculated using ground truth labels to help in model tuning, evaluation, and selection. In an unsupervised setting, however, where the data has no ground truth, there are few interpretable metrics that can guide us to do the same. Our approach creates a framework to adapt the Intersection over Union metric, referred to as Adapted IoU, usually used to evaluate supervised learning models, into the unsupervised domain, which solves the problem by factoring in subject matter expertise and intuition about the ideal output from the model. This metric essentially provides a scale that allows us to compare the performance across numerous unsupervised models or tune hyper-parameters and compare different versions of the same model.

Keywords: general metric, unsupervised learning, classification, intersection over union

Procedia PDF Downloads 28

Authors: Mohammad Alnabhan, Awni Hammouri, Mustafa Hammad, Anas Al-Badareen, Omamah Al-Thnebat

Abstract:

Software visualization is one of the main techniques used to simplify the presentation of software systems and enhance their understandability. It is used to present the software system in a visual manner using simple, clear and meaningful symbols. This study proposes a new 2D software visualization approach. In this approach, each class is represented by rectangle, the name of the class placed above the rectangle, the size of class (Line of Code) represented by the height of the rectangle. The methods and the attributes are represented by circles and triangles respectively. The relationships among classes correspond to arrows. The proposed visualization approach was evaluated in terms of applicability and efficiency. Results have confirmed successful implementation of the proposed approach, and its ability to provide a simple and effective graphical presentation of extracted software components and properties.

Keywords: software visualization, software metrics, calling relationships, 2D graphs

Procedia PDF Downloads 179

Authors: T. Francis Thamburaj, A. Aloysius

Abstract:

Polymorphism is one of the main pillars of the object-oriented paradigm. It induces hidden forms of class dependencies which may impact software quality, resulting in higher cost factor for comprehending, debugging, testing, and maintaining the software. In this paper, a new cognitive complexity metric called Cognitive Weighted Polymorphism Factor (CWPF) is proposed. Apart from the software structural complexity, it includes the cognitive complexity on the basis of type. The cognitive weights are calibrated based on 27 empirical studies with 120 persons. A case study and experimentation of the new software metric shows positive results. Further, a comparative study is made and the correlation test has proved that CWPF complexity metric is a better, more comprehensive, and more realistic indicator of the software complexity than Abreu’s Polymorphism Factor (PF) complexity metric.

Keywords: cognitive complexity metric, object-oriented metrics, polymorphism factor, software metrics

Procedia PDF Downloads 431

Authors: Maryam Amiri, Leyli Mohammad-Khanli

Abstract:

Cloud computing is becoming increasingly popular and more business applications are moving to cloud. In this regard, services that provide similar functional properties are increasing. So, the ability to select a service with the best non-functional properties, corresponding to the user preference, is necessary for the user. This paper presents an Evaluation Framework of Service Trustworthiness (EFST) that evaluates the trustworthiness of equivalent services without need to additional invocations of them. EFST extracts user preference automatically. Then, it assesses trustworthiness of services in two dimensions of qualitative and quantitative metrics based on the experiences of past usage of services. Finally, EFST determines the overall trustworthiness of services using Fuzzy Inference System (FIS). The results of experiments and simulations show that EFST is able to predict the missing values of Quality of Service (QoS) better than other competing approaches. Also, it propels users to select the most appropriate services.

Keywords: user preference, cloud service, trustworthiness, QoS metrics, prediction

Procedia PDF Downloads 268

Authors: Zhaoqi He

Abstract:

Research Context: Endurance performance in hot environments is influenced by the interplay of hydration status and physiological responses. This study aims to investigate how dehydration, up to 2.11% body weight loss, affects the 3 km running performance of trained male athletes under conditions mimicking high temperatures. Methodology: In a randomized crossover design, five male athletes participated in two trials – euhydrated (EU) and dehydrated (HYPO). Both trials included a 70-minute preload run at 55-60% VO2max in 32°C and 50% humidity, followed by a 3-kilometer time trial. Fluid intake was restricted in HYPO to induce a 2.11% body weight loss. Physiological metrics, including heart rate, core temperature, and oxygen uptake, were measured, along with perceptual metrics like perceived exertion and thirst sensation. Findings: The 3-kilometer run completion times showed no significant differences between EU and HYPO trials (p=0.944). Physiological indicators, including heart rate, core temperature, and oxygen uptake, did not significantly vary (p>0.05). Thirst sensation was markedly higher in HYPO (p=0.013), confirming successful induction of dehydration. Other perceptual metrics and gastrointestinal comfort remained consistent. Conclusion: Contrary to the hypothesis, the study reveals that dehydration, inducing up to 2.11% body weight loss, does not significantly impair 3 km running performance in trained male athletes under hot conditions. Thirst sensation was notably higher in the dehydrated state, emphasizing the importance of considering perceptual factors in hydration strategies. The findings suggest that trained runners can maintain performance despite moderate dehydration, highlighting the need for nuanced hydration guidelines in hot-weather running.

Keywords: hypohydration, euhydration, hot environment, 3km running time trial, endurance performance, trained athletes, perceptual metrics, dehydration impact, physiological responses, hydration strategies

Procedia PDF Downloads 40

Authors: Volkan Uygun, H. Birkan Yilmaz, Tuna Tugcu

Abstract:

This study focuses on the development and use of algorithms that address the issue of resource management in response to emergency and disaster situations. The presented system, named Disaster Management Platform (DMP), takes the data from the data sources of service providers and distributes the incoming requests accordingly both to manage load balancing and minimize service time, which results in improved user satisfaction. Three different resource management algorithms, which give different levels of importance to load balancing and service time, are proposed for the study. The first one is the Minimum Distance algorithm, which assigns the request to the closest resource. The second one is the Minimum Load algorithm, which assigns the request to the resource with the minimum load. Finally, the last one is the Hybrid algorithm, which combines the previous two approaches. The performance of the proposed algorithms is evaluated with respect to waiting time, success ratio, and maximum load ratio. The metrics are monitored from simulations, to find the optimal scheme for different loads. Two different simulations are performed in the study, one is time-based and the other is lambda-based. The results indicate that, the Minimum Load algorithm is generally the best in all metrics whereas the Minimum Distance algorithm is the worst in all cases and in all metrics. The leading position in performance is switched between the Minimum Distance and the Hybrid algorithms, as lambda values change.

Keywords: emergency and disaster response, resource management algorithm, disaster situations, disaster management platform

Procedia PDF Downloads 323

Authors: Ana Sofia Torres Pereira

Abstract:

Till this day, women have been underrepresented and stereotyped both in TV and Cinema Screens all around the World. While women have been gaining a different status and finding their own voice in the work place and in society, what we see on-screen is still something different, something gender biased, something that does not show the multifaceted identities a woman might have. But why is this so? Why are we stuck on this shallow vision of women on-screen? According to several cinema industry studies, most film screenwriters in Hollywood are men. Women actually represent a very low percentage of screenwriters. So why is this relevant? Could the underrepresentation of women screenwriters in Hollywood be affecting the way women are written, and as a result, are depicted in film? Films are about stories, about people, and if these stories are continuously told through a man’s gaze, is that helping in the creation of a gender imbalance towards women? On the other hand, one of the reasons given for the low percentage of women screenwriters is: women are said to be better at writing specific genres, like dramas and comedies, and not as good writing thrillers and action films, so, as women seem to be limited in the genres they can write, they are undervalued and underrepresented as screenwriters. It seems the gender bias and stereotype isn’t saved exclusively for women on-screen, but also off-screen and behind the screen. So film appears to be a men’s world, on and off-screen, and since men seem to write the majority of scripts, it might be no wonder that women have been written in a specific way and depicted in a specific way on-screen. Also, since films are a mass communication medium, maybe this over-sexualization and stereotyping on-screen is indoctrinating our society into believing this bias is alive and well, and thus targeting women off-screen as well (ergo, screenwriters). What about at the very begging of film? In the Silent Movies and Early Talkies era, women dominated the screenwriting industry. They wrote every genre, and the majority of scripts were written by women, not men. So what about then? How were women depicted in films then? Did women screenwriters, in an era that was still very harsh on women, use their stories and their power to break stereotypes and show women in a different light, or did they carry on with the stereotype, did they continue it and standardize it? This papers aims to understand how important it is to have more working women screenwriters in order to break stereotypes regarding the image of women on and off-screen. How much can a screenwriter (male or female) influence our gaze on women (on and off-screen)?

Keywords: cinema, gender bias, stereotype, women on-screen, women screenwriters

Procedia PDF Downloads 329

Authors: Mohammadali Tavakoli, Atefeh Heydari, Zuriati Ismail, Naomie Salim

Abstract:

With the increasing number of people reviewing products online in recent years, opinion sharing websites has become the most important source of customers’ opinions. Unfortunately, spammers generate and post fake reviews in order to promote or demote brands and mislead potential customers. These are notably destructive not only for potential customers but also for business holders and manufacturers. However, research in this area is not adequate, and many critical problems related to spam detection have not been solved to date. To provide green researchers in the domain with a great aid, in this paper, we have attempted to create a high-quality framework to make a clear vision on review spam-detection methods. In addition, this report contains a comprehensive collection of detection metrics used in proposed spam-detection approaches. These metrics are extremely applicable for developing novel detection methods.

Keywords: fake reviews, feature collection, opinion spam, spam detection

Procedia PDF Downloads 393

Authors: Hadi Badri Ahmadi, Xuping Wang

Abstract:

Due to strict regulations and public awareness, corporations should develop policies to effectively decrease the negative environmental effects of their products and enhance their supply chain environmental sustainability. Assessment of environmental issues in the context of many industries has been studied in the previous literature. However, Iran cement industry has received less attention from researchers. Therefore, in this paper, we apply a Decision-Making Trial and Evaluation Laboratory (DEMATEL) approach to assess the relationships among green metrics of Iran cement industry supply chain under fuzzy environment. The study findings provide considerable insight for cement industry managers and experts in order to enhance the environmental sustainability of their supply chain and move towards sustainable development.

Keywords: green supply chain, DEMATEL, fuzzy set theory, environmental sustainability, sustainable development, cement industry

Procedia PDF Downloads 388

Authors: Cesar Hernández, Diego Giral, Ingrid Páez

Abstract:

This article proposes a hybrid algorithm for spectrum allocation in cognitive radio networks based on the algorithms Analytical Hierarchical Process (AHP) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to improve the performance of the spectrum mobility of secondary users in cognitive radio networks. To calculate the level of performance of the proposed algorithm a comparative analysis between the proposed AHP-TOPSIS, Grey Relational Analysis (GRA) and Multiplicative Exponent Weighting (MEW) algorithm is performed. Four evaluation metrics is used. These metrics are the accumulative average of failed handoffs, the accumulative average of handoffs performed, the accumulative average of transmission bandwidth, and the accumulative average of the transmission delay. The results of the comparison show that AHP-TOPSIS Algorithm provides 2.4 times better performance compared to a GRA Algorithm and, 1.5 times better than the MEW Algorithm.

Keywords: cognitive radio, decision making, hybrid algorithm, spectrum handoff, wireless networks

Procedia PDF Downloads 520

Authors: K. Hema Shankari, R. Thirumalaiselvi, N. V. Balasubramanian

Abstract:

The present paper addresses to the research in the area of regression testing with emphasis on automated tools as well as prioritization of test cases. The uniqueness of regression testing and its cyclic nature is pointed out. The difference in approach between industry, with business model as basis, and academia, with focus on data mining, is highlighted. Test Metrics are discussed as a prelude to our formula for prioritization; a case study is further discussed to illustrate this methodology. An industrial case study is also described in the paper, where the number of test cases is so large that they have to be grouped as Test Suites. In such situations, a genetic algorithm proposed by us can be used to reconfigure these Test Suites in each cycle of regression testing. The comparison is made between a proprietary tool and an open source tool using the above-mentioned metrics. Our approach is clarified through several tables.

Keywords: APFD metric, genetic algorithm, regression testing, RFT tool, test case prioritization, selenium tool

Procedia PDF Downloads 412

Authors: Shareefa Abdullah Al-Maqtari, Ruzaika Omar Basaree, Rafeah Legino

Abstract:

A pictorial representation of hybrid forms in science-art collaboration has become a crucial issue in the course of exploring a new painting technique development. This is straight related to the reception of an invisible-recognition phenomenology. In hybrid pictorial representation of invisible-recognition phenomenology, the challenging issue is how to depict the pictorial features of indescribable objects from its mental source, modality and transparency. This paper proposes the hybrid technique of painting Demonstrate, Resemble, and Synthesize (DRS) through a combination of the hybrid aspect-recognition representation of understanding picture, demonstrative mod, the number theory, pattern in the modular arithmetic system, and the coherence theory of visual attention in the dynamic scenes representation. Multi-methods digital gaze data analyses, pattern-modular table operation design, and rotation parameter were used for the visualization. In the scientific processes, Eye-trackingvideo-sections based was conducted using Tobii T60 remote eye tracking hardware and TobiiStudioTM analysis software to collect and analyze the eye movements of ten participants when watching the video clip, Alexander Paulikevitch’s performance’s ‘Tajwal’. Results: we found that correlation of fixation count in section one was positively and moderately correlated with section two Person’s (r=.10, p < .05, 2-tailed) as well as in fixation duration Person’s (r=.10, p < .05, 2-tailed). However, a paired-samples t-test indicates that scores were significantly higher for the section one (M = 2.2, SD = .6) than for the section two (M = 1.93, SD = .6) t(9) = 2.44, p < .05, d = 0.87. In the visual process, the exported data of gaze number N was resembled the hybrid forms of visuospatial attention using the table-mod-analyses operation. The explored hybrid guideline was simply applicable, and it could be as alternative approach to the sustainability of contemporary visual arts.

Keywords: science-art collaboration, hybrid forms, pictorial representation, visuospatial attention, modular arithmetic

Procedia PDF Downloads 349

Authors: Cesar Hernández, Diego Giral, Fernando Santa

Abstract:

The spectral handoff is important in cognitive wireless networks to ensure an adequate quality of service and performance for secondary user communications. This work proposes a benchmarking of performance of the three spectrum handoff models: VIKOR, SAW and MEW. Four evaluation metrics are used. These metrics are, accumulative average of failed handoffs, accumulative average of handoffs performed, accumulative average of transmission bandwidth and, accumulative average of the transmission delay. As a difference with related work, the performance of the three spectrum handoff models was validated with captured data of spectral occupancy in experiments realized at the GSM frequency band (824 MHz-849 MHz). These data represent the actual behavior of the licensed users for this wireless frequency band. The results of the comparative show that VIKOR Algorithm provides 15.8% performance improvement compared to a SAW Algorithm and, 12.1% better than the MEW Algorithm.

Keywords: cognitive radio, decision making, MEW, SAW, spectrum handoff, VIKOR

Procedia PDF Downloads 417

Authors: Martin H¨ausl, Maximilian Auch, Johannes Forster, Peter Mandl, Alexander Schill

Abstract:

In order to survive on the market, companies must constantly develop improved and new products. These products are designed to serve the needs of their customers in the best possible way. The creation of new products is also called innovation and is primarily driven by a company’s internal research and development department. However, a new approach has been taking place for some years now, involving external knowledge in the innovation process. This approach is called open innovation and identifies customer knowledge as the most important source in the innovation process. This paper presents a concept of using social media posts as an external source to support the open innovation approach in its initial phase, the Ideation phase. For this purpose, the social media posts are semantically structured with the help of an ontology and the authors are evaluated using graph-theoretical metrics such as density. For the structuring and evaluation of relevant social media posts, we also use the findings of Natural Language Processing, e. g. Named Entity Recognition, specific dictionaries, Triple Tagger and Part-of-Speech-Tagger. The selection and evaluation of the tools used are discussed in this paper. Using our ontology and metrics to structure social media posts enables users to semantically search these posts for new product ideas and thus gain an improved insight into the external sources such as customer needs.

Keywords: idea ontology, innovation management, semantic search, open information extraction

Procedia PDF Downloads 172

Authors: Doyin Afolabi, Phillip Adewole, Oladipupo Sennaike

Abstract:

Most well-known classifiers, including the decision tree algorithm, can make predictions on balanced datasets efficiently. However, the decision tree algorithm tends to be biased towards imbalanced datasets because of the skewness of the distribution of such datasets. To overcome this problem, this study proposes a weighted decision tree algorithm that aims to remove the bias toward the majority class and prevents the reduction of majority observations in imbalance datasets classification. The proposed weighted decision tree algorithm was tested on three imbalanced datasets- cancer dataset, german credit dataset, and banknote dataset. The specificity, sensitivity, and accuracy metrics were used to evaluate the performance of the proposed decision tree algorithm on the datasets. The evaluation results show that for some of the weights of our proposed decision tree, the specificity, sensitivity, and accuracy metrics gave better results compared to that of the ID3 decision tree and decision tree induced with minority entropy for all three datasets.

Keywords: data mining, decision tree, classification, imbalance dataset

Procedia PDF Downloads 108

Authors: Jehad Al Dallal

Abstract:

The design of the internal relationships among object-oriented class members (i.e., attributes and methods) and the external relationships among classes affects the overall quality of the object-oriented software. The degree of relatedness among class members is referred to as class cohesion and the degree to which a class is related to other classes is called class coupling. Well designed classes are expected to exhibit high cohesion and low coupling values. In this paper, using classes of three open-source Java systems, we empirically investigate the relation between class cohesion and coupling. In the empirical study, five lack-of-cohesion metrics and eight coupling metrics are considered. The empirical study results show that class cohesion and coupling internal quality attributes are inversely correlated. The strength of the correlation highly depends on the cohesion and coupling measurement approaches.

Keywords: class cohesion measure, class coupling measure, object-oriented class, software quality

Procedia PDF Downloads 216

Authors: Siddhartha Khare, Suyash Khare

Abstract:

Phenology, a crucial discipline in ecology, offers insights into the seasonal dynamics of organisms within natural ecosystems and the underlying environmental triggers. Leveraging the potent capabilities of digital repeat photography, PhenoCams capture invaluable data on the phenology of crops, plants, and trees. These cameras yield digital imagery in Red Green Blue (RGB) color channels, and some advanced systems even incorporate Near Infrared (NIR) bands. This study presents compelling case studies employing PhenoCam technology to unravel the phenology of black spruce trees. Through the analysis of RGB color channels, a range of essential color metrics including red chromatic coordinate (RCC), green chromatic coordinate (GCC), blue chromatic coordinate (BCC), vegetation contrast index (VCI), and excess green index (ExGI) are derived. These metrics illuminate variations in canopy color across seasons, shedding light on bud and leaf development. This, in turn, facilitates a deeper understanding of phenological events and aids in delineating the growth periods of trees and plants. The initial phase of this study addresses critical questions surrounding the fidelity of continuous canopy greenness records in representing bud developmental phases. Additionally, it discerns which color-based index most accurately tracks the seasonal variations in tree phenology within evergreen forest ecosystems. The subsequent section of this study delves into the transition dates of black spruce (Picea mariana (Mill.) B.S.P.) phenology. This is achieved through a fortnightly comparative analysis of the MODIS normalized difference vegetation index (NDVI) and the enhanced vegetation index (EVI). By employing PhenoCam technology and leveraging advanced color metrics, this study significantly advances our comprehension of black spruce tree phenology, offering valuable insights for ecological research and management.

Keywords: phenology, remote sensing, phenocam, color metrics, NDVI, GCC

Procedia PDF Downloads 43

Authors: Fouzi Harrou, Ying Sun, Sofiane Khadraoui

Abstract:

Anomaly detection based on Principal Component Analysis (PCA) was studied intensively and largely applied to multivariate processes with highly cross-correlated process variables. Monitoring metrics such as the Hotelling's T2 and the Q statistics are usually used in PCA-based monitoring to elucidate the pattern variations in the principal and residual subspaces, respectively. However, these metrics are ill suited to detect small faults. In this paper, the Exponentially Weighted Moving Average (EWMA) based on the Q and T statistics, T2-EWMA and Q-EWMA, were developed for detecting faults in the process mean. The performance of the proposed methods was compared with that of the conventional PCA-based fault detection method using synthetic data. The results clearly show the benefit and the effectiveness of the proposed methods over the conventional PCA method, especially for detecting small faults in highly correlated multivariate data.

Keywords: data-driven method, process control, anomaly detection, dimensionality reduction

Procedia PDF Downloads 276

Authors: Debarag Banerjee, Pooja Singh, Arjun Avadhanam, Saksham Srivastava

Abstract:

Autonomous conversational agents, i.e., chatbots, are becoming an increasingly common mechanism for enterprises to provide support to customers and partners. In order to rate chatbots, especially ones powered by Generative AI tools like Large Language Models (LLMs), we need to be able to accurately assess their performance. This is where chatbot benchmarking becomes important. In this paper, authors propose the use of a benchmark that they call the E2E (End to End) benchmark and show how the E2E benchmark can be used to evaluate the accuracy and usefulness of the answers provided by chatbots, especially ones powered by LLMs. The authors evaluate an example chatbot at different levels of sophistication based on both our E2E benchmark as well as other available metrics commonly used in the state of the art and observe that the proposed benchmark shows better results compared to others. In addition, while some metrics proved to be unpredictable, the metric associated with the E2E benchmark, which uses cosine similarity, performed well in evaluating chatbots. The performance of our best models shows that there are several benefits of using the cosine similarity score as a metric in the E2E benchmark.

Keywords: chatbot benchmarking, end-to-end (E2E) benchmarking, large language model, user centric evaluation.

Procedia PDF Downloads 49

Authors: Elahe Kamari

Abstract:

Research on translation process has been conducted for more than 20 years, investigating various issues and using different research methodologies. Most recently, researchers have started to use eye tracking to study translation processes. They believed that the observable, measurable data that can be gained from eye tracking are indicators of unobservable cognitive processes happening in the translators’ mind during translation tasks. The aim of this study was to investigate directionality in translation processes through using eye tracking. The following hypotheses were tested: 1) processing the target text requires more cognitive effort than processing the source text, in both directions of translation; 2) L2 translation tasks on the whole require more cognitive effort than L1 tasks; 3) cognitive resources allocated to the processing of the source text is higher in L1 translation than in L2 translation; 4) cognitive resources allocated to the processing of the target text is higher in L2 translation than in L1 translation; and 5) in both directions non-professional translators invest more cognitive effort in translation tasks than do professional translators. The performance of a group of 30 male professional translators was compared with that of a group of 30 male non-professional translators. All the participants translated two comparable texts one into their L1 (Persian) and the other into their L2 (English). The eye tracker measured gaze time, average fixation duration, total task length and pupil dilation. These variables are assumed to measure the cognitive effort allocated to the translation task. The data derived from eye tracking only confirmed the first hypothesis. This hypothesis was confirmed by all the relevant indicators: gaze time, average fixation duration and pupil dilation. The second hypothesis that L2 translation tasks requires allocation of more cognitive resources than L1 translation tasks has not been confirmed by all four indicators. The third hypothesis that source text processing requires more cognitive resources in L1 translation than in L2 translation and the fourth hypothesis that target text processing requires more cognitive effort in L2 translation than L1 translation were not confirmed. It seems that source text processing in L2 translation can be just as demanding as in L1 translation. The final hypothesis that non-professional translators allocate more cognitive resources for the same translation tasks than do the professionals was partially confirmed. One of the indicators, average fixation duration, indicated higher cognitive effort-related values for professionals.

Keywords: translation processes, eye tracking, cognitive resources, directionality

Procedia PDF Downloads 443

Authors: Sezin Ekşioğlu

Abstract:

Multilabel classification has a huge importance for several applications, it is also a challenging research topic. It is a kind of supervised learning that contains binary targets. The distance between multilabel and binary classification is having more than one class in multilabel classification problems. Features can belong to one class or many classes. There exists a wide range of applications for multi label prediction such as image labeling, text categorization, gene functionality. Even though features are classified in many classes, they may not always be properly classified. There are many ensemble methods for the classification. However, most of the researchers have been concerned about better multilabel methods. Especially little ones focus on both efficiency of classifiers and pairwise relationships at the same time in order to implement better multilabel classification. In this paper, we worked on modified ensemble methods by getting benefit from k-Nearest Neighbors and neural network structure to address issues within a beneficial way and to get better impacts from the multilabel classification. Publicly available datasets (yeast, emotion, scene and birds) are performed to demonstrate the developed algorithm efficiency and the technique is measured by accuracy, F1 score and hamming loss metrics. Our algorithm boosts benchmarks for each datasets with different metrics.

Keywords: multilabel, classification, neural network, KNN

Procedia PDF Downloads 136

Authors: Arvind Salem

Abstract:

Voters across the country are transferring the power of redistricting from the state legislatures to commissions to secure “fairer” districts by curbing the influence of gerrymandering on redistricting. Gerrymandering, intentionally drawing distorted districts to achieve political advantage, has become extremely prevalent, generating widespread voter dissatisfaction and resulting in states adopting commissions for redistricting. However, the efficacy of these commissions is dubious, with some arguing that they constitute a panacea for gerrymandering, while others contend that commissions have relatively little effect on gerrymandering. A result showing that commissions are effective would allay these fears, supplying ammunition for activists across the country to advocate for commissions in their state and reducing the influence of gerrymandering across the nation. However, a result against commissions may reaffirm doubts about commissions and pressure lawmakers to make improvements to commissions or even abandon the commission system entirely. Additionally, these commissions are publicly funded: so voters have a financial interest and responsibility to know if these commissions are effective. Currently, nine states place commissions in charge of redistricting, Arizona, California, Colorado, Michigan, Idaho, Montana, Washington, and New Jersey (Hawaii also has a commission but will be excluded for reasons mentioned later). This study compares the degree of gerrymandering in the 2022 election (“after”) to the election in which voters decided to adopt commissions (“before”). The before-election provides a valuable benchmark for assessing the efficacy of commissions since voters in those elections clearly found the districts to be unfair; therefore, comparing the current election to that one is a good way to determine if commissions have improved the situation. At the time Hawaii adopted commissions, it was merely a single at-large district, so it is before metrics could not be calculated, and it was excluded. This study will use three methods to quantify the degree of gerrymandering: the efficiency gap, the percentage of seats and the percentage of votes difference, and the mean-median difference. Each of these metrics has unique advantages and disadvantages, but together, they form a balanced approach to quantifying gerrymandering. The study uses a Wilcoxon Signed-Rank Test with a null hypothesis that the value of the metrics is greater than or equal to after the election than before and an alternative hypothesis that the value of these metrics is greater in the before the election than after using a 0.05 significance level and an expected difference of 0. Accepting the alternative hypothesis would constitute evidence that commissions reduce gerrymandering to a statistically significant degree. However, this study could not conclude that commissions are effective. The p values obtained for all three metrics (p=0.42 for the efficiency gap, p=0.94 for the percentage of seats and percentage of votes difference, and p=0.47 for the mean-median difference) were extremely high and far from the necessary value needed to conclude that commissions are effective. These results halt optimism about commissions and should spur serious discussion about the effectiveness of these commissions and ways to change them moving forward so that they can accomplish their goal of generating fairer districts.

Keywords: commissions, elections, gerrymandering, redistricting

Procedia PDF Downloads 60

Authors: Heshalini Rajagopal, Li Sze Chow, Raveendran Paramesran

Abstract:

Magnetic Resonance Imaging (MRI) is one of the most important medical imaging modality. Subjective assessment of the image quality is regarded as the gold standard to evaluate MR images. In this study, a database of 210 MR images which contains ten reference images and 200 distorted images is presented. The reference images were distorted with four types of distortions: Rician Noise, Gaussian White Noise, Gaussian Blur and DCT compression. The 210 images were assessed by ten subjects. The subjective scores were presented in Difference Mean Opinion Score (DMOS). The DMOS values were compared with four FR-IQA metrics. We have used Pearson Linear Coefficient (PLCC) and Spearman Rank Order Correlation Coefficient (SROCC) to validate the DMOS values. The high correlation values of PLCC and SROCC shows that the DMOS values are close to the objective FR-IQA metrics.

Keywords: medical resonance (MR) images, difference mean opinion score (DMOS), full reference image quality assessment (FR-IQA)

Procedia PDF Downloads 445

Authors: Wuttigrai Ngamsirijit

Abstract:

Talent management in today’s modern organizations has become data-driven due to a demand for objective human resource decision making and development of analytics technologies. HR managers have been faced with some obstacles in exploiting data and information to obtain their effective talent management decisions. These include process-based data and records; insufficient human capital-related measures and metrics; lack of capabilities in data modeling in strategic manners; and, time consuming to add up numbers and make decisions. This paper proposes a framework of talent management through integration of talent value chain and human capital analytics approaches. It encompasses key data, measures, and metrics regarding strategic talent management decisions along the organizational and talent value chain. Moreover, specific predictive and prescriptive models incorporating these data and information are recommended to help managers in understanding the state of talent, gaps in managing talent and the organization, and the ways to develop optimized talent strategies.    

Keywords: decision making, human capital analytics, talent management, talent value chain

Procedia PDF Downloads 158

Authors: Aviv Yaniv, Ron Ben Arosh, Nadav Gasner, Michael Konviser, Arbel Yaniv

Abstract:

This paper introduces two methods for the evaluation of Question Decomposition Meaning Representation (QDMR) as predicted by sequence-to-sequence model and COPYNET parser for natural language questions processing, motivated by the fact that previous evaluation metrics used for this task do not take into account some characteristics of the representation, such as partial ordering structure. To this end, several heuristics to extract such partial dependencies are formulated, followed by the hereby proposed evaluation methods denoted as Proportional Graph Matcher (PGM) and Conversion to Normal String Representation (Nor-Str), designed to better capture the accuracy level of QDMR predictions. Experiments are conducted to demonstrate the efficacy of the proposed evaluation methods and show the added value suggested by one of them- the Nor-Str, for better distinguishing between high and low-quality QDMR when predicted by models such as COPYNET. This work represents an important step forward in the development of better evaluation methods for QDMR predictions, which will be critical for improving the accuracy and reliability of natural language question-answering systems.

Keywords: NLP, question answering, question decomposition meaning representation, QDMR evaluation metrics

Procedia PDF Downloads 55

Authors: Larissa Pliska, Isabel Neitzel, Michael Buschermöhle, Olga Kunina-Habenicht, Ute Ritterfeld

Abstract:

Visual preferences, which can be assessed using eye tracking technologies, are considered one of the defining hallmarks of Autism Spectrum Disorder (ASD). Specifically, children with ASD show a decreased preference for social images rather than geometric images compared to typically developed (TD) children. Such differences are already prevalent at a very early age and indicate the severity of the disorder: toddlers with ASD who preferred geometric images when confronted with social and geometric images showed higher ASD symptom severity than toddlers with ASD who showed higher social attention. Furthermore, the complexity of social pictures (one child playing vs. two children playing together) as well as the mode of stimulus presentation (video or image), are not decisive for the marker. The average age of diagnosis for ASD in Germany is 6.5 years, and visual preference data on this age group is missing. In the present study, we therefore investigated whether visual preferences persist into school age. We examined the visual preferences of 16 boys aged 6 to 11 with ASD and unimpaired cognition as well as TD children (1:1 matching based on children's age and the parent's level of education) within an experimental setting. Different stimulus presentation formats (images vs. videos) and different levels of stimulus complexity were included. Children with and without ASD received pairs of social and non-social images and video stimuli on a screen while eye movements (i.e., eye position and gaze direction) were recorded. For this specific use case, KIZMO GmbH developed a customized, native iOS app (KIZMO Face-Analyzer) for use on iPads. Neither the format of stimulus presentation nor the complexity of the social images had a significant effect on the visual preference of children with and without ASD in this study. Despite the tendency for a difference between the groups for the video stimuli, there were no significant differences. Overall, no statistical differences in visual preference occurred between boys with and without ASD, suggesting that gaze preference in these groups is similar at primary school age. One limitation is that the children with ASD were already receiving Autism-specific intervention. The potential of a visual preference task as an indicator of ASD can be emphasized. The article discusses the clinical relevance of this marker in elementary school children.

Keywords: autism spectrum disorder, eye tracking, hallmark, visual preference

Procedia PDF Downloads 39

Authors: M. Velasco, D. While, P. Raju, J. Krasniewicz, A. Amini, L. Hernandez-Munoz

Abstract:

Web service adaptation involves the creation of adapters that solve Web services incompatibilities known as mismatches. Since the importance of Web services adaptation is increasing because of the frequent implementation and use of online Web services, this paper presents a literature review of web services to investigate the main methods of adaptation, their theoretical underpinnings and the metrics used to measure adapters performance. Eighteen publications were reviewed independently by two researchers. We found that adaptation techniques are needed to solve different types of problems that may arise due to incompatibilities in Web service interfaces, including protocols, messages, data and semantics that affect the interoperability of the services. Although adapters are non-invasive methods that can improve Web services interoperability and there are current approaches for service adaptation; there is, however, not yet one solution that fits all types of mismatches. Our results also show that only a few research projects incorporate theoretical frameworks and that metrics to measure adapters’ performance are very limited. We conclude that further research on software adaptation should improve current adaptation methods in different layers of the service interoperability and that an adaptation theoretical framework that incorporates a theoretical underpinning and measures of qualitative and quantitative performance needs to be created.

Keywords: Web Services Adapters, software adaptation, web services mismatches, web services interoperability

Procedia PDF Downloads 272

Authors: Faishal Muhammad, Erlinda Muslim, Nadia Faradilla, Sayidul Fikri

Abstract:

To deal with the increase of mass transportation needs problem, PT. Kereta Commuter Jabodetabek (KCJ) implements Commuter Vending Machine (C-VIM) as the solution. For that background, C-VIM is implemented as a substitute to the conventional ticket windows with the purposes to make transaction process more efficient and to introduce self-service technology to the commuter line user. However, this implementation causing problems and long queues when the user is not accustomed to using the machine. The objective of this research is to evaluate user experience after using the commuter vending machine. The goal is to analyze the existing user experience problem and to achieve a better user experience design. The evaluation method is done by giving task scenario according to the features offered by the machine. The features are daily insured ticket sales, ticket refund, and multi-trip card top up. There 20 peoples that separated into two groups of respondents involved in this research, which consist of 5 males and 5 females each group. The experienced and inexperienced user to prove that there is a significant difference between both groups in the measurement. The user experience is measured by both quantitative and qualitative measurement. The quantitative measurement includes the user performance metrics such as task success, time on task, error, efficiency, and learnability. The qualitative measurement includes system usability scale questionnaire (SUS), questionnaire for user interface satisfaction (QUIS), and retrospective think aloud (RTA). Usability performance metrics shows that 4 out of 5 indicators are significantly different in both group. This shows that the inexperienced group is having a problem when using the C-VIM. Conventional ticket windows also show a better usability performance metrics compared to the C-VIM. From the data processing, the experienced group give the SUS score of 62 with the acceptability scale of 'marginal low', grade scale of “D”, and the adjective ratings of 'good' while the inexperienced group gives the SUS score of 51 with the acceptability scale of 'marginal low', grade scale of 'F', and the adjective ratings of 'ok'. This shows that both groups give a low score on the system usability scale. The QUIS score of the experienced group is 69,18 and the inexperienced group is 64,20. This shows the average QUIS score below 70 which indicate a problem with the user interface. RTA was done to obtain user experience issue when using C-VIM through interview protocols. The issue obtained then sorted using pareto concept and diagram. The solution of this research is interface redesign using activity relationship chart. This method resulted in a better interface with an average SUS score of 72,25, with the acceptable scale of 'acceptable', grade scale of 'B', and the adjective ratings of 'excellent'. From the time on task indicator of performance metrics also shows a significant better time by using the new interface design. Result in this study shows that C-VIM not yet have a good performance and user experience.

Keywords: activity relationship chart, commuter line vending machine, system usability scale, usability performance metrics, user experience evaluation

Procedia PDF Downloads 245

Authors: Justin Reverdi, Sixin Zhang, Saïd Aoues, Fabrice Gamboa, Serge Gratton, Thomas Pellegrini

Abstract:

In vapor cycle systems, the mass flow sensor plays a key role for different monitoring and control purposes. However, physical sensors can be inaccurate, heavy, cumbersome, expensive, or highly sensitive to vibrations, which is especially problematic when embedded into an aircraft. The conception of a virtual sensor, based on other standard sensors, is a good alternative. This paper has two main objectives. Firstly, a data-driven model using a convolutional neural network is proposed to estimate the mass flow of the compressor. We show that it significantly outperforms the standard polynomial regression model (thermodynamic maps) in terms of the standard MSE metric and engineer performance metrics. Secondly, a semi-automatic segmentation method is proposed to compute the engineer performance metrics for real datasets, as the standard MSE metric may pose risks in analyzing the dynamic behavior of vapor cycle systems.

Keywords: deep learning, convolutional neural network, vapor cycle system, virtual sensor

Procedia PDF Downloads 39

Authors: Alireza Vafaei Sadr, Vida Abedi, Jiang Li, Ramin Zand

Abstract:

Missing data is a common challenge in medical research and can lead to biased or incomplete results. When the data bias leaks into models, it further exacerbates health disparities; biased algorithms can lead to misclassification and reduced resource allocation and monitoring as part of prevention strategies for certain minorities and vulnerable segments of patient populations, which in turn further reduce data footprint from the same population – thus, a vicious cycle. This study compares the performance of six imputation techniques grouped into Linear and Non-Linear models on two different realworld electronic health records (EHRs) datasets, representing 17864 patient records. The mean absolute percentage error (MAPE) and root mean squared error (RMSE) are used as performance metrics, and the results show that the Linear models outperformed the Non-Linear models in terms of both metrics. These results suggest that sometimes Linear models might be an optimal choice for imputation in laboratory variables in terms of imputation efficiency and uncertainty of predicted values.

Keywords: EHR, machine learning, imputation, laboratory variables, algorithmic bias

Procedia PDF Downloads 67