Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00090
Guojing Cong, Talia Ben-Naim, Victor Fung, Anshul Gupta, R. Neumann, Mathias Steiner
We present our research where attention mechanism is extensively applied to various aspects of graph neural net- works for predicting materials properties. As a result, surrogate models can not only replace costly simulations for materials screening but also formulate hypotheses and insights to guide further design exploration. We predict formation energy of the Materials Project and gas adsorption of crystalline adsorbents, and demonstrate the superior performance of our graph neural networks. Moreover, attention reveals important substructures that the machine learning models deem important for a material to achieve desired target properties. Our model is based solely on standard structural input files containing atomistic descriptions of the adsorbent material candidates. We construct novel methodological extensions to match the prediction accuracy of state-of-the-art models some of which were built with hundreds of features at much higher computational cost. We show that sophisticated neural networks can obviate the need for elaborate feature engineering. Our approach can be more broadly applied to optimize gas capture processes at industrial scale.
{"title":"Extensive Attention Mechanisms in Graph Neural Networks for Materials Discovery","authors":"Guojing Cong, Talia Ben-Naim, Victor Fung, Anshul Gupta, R. Neumann, Mathias Steiner","doi":"10.1109/ICDMW58026.2022.00090","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00090","url":null,"abstract":"We present our research where attention mechanism is extensively applied to various aspects of graph neural net- works for predicting materials properties. As a result, surrogate models can not only replace costly simulations for materials screening but also formulate hypotheses and insights to guide further design exploration. We predict formation energy of the Materials Project and gas adsorption of crystalline adsorbents, and demonstrate the superior performance of our graph neural networks. Moreover, attention reveals important substructures that the machine learning models deem important for a material to achieve desired target properties. Our model is based solely on standard structural input files containing atomistic descriptions of the adsorbent material candidates. We construct novel methodological extensions to match the prediction accuracy of state-of-the-art models some of which were built with hundreds of features at much higher computational cost. We show that sophisticated neural networks can obviate the need for elaborate feature engineering. Our approach can be more broadly applied to optimize gas capture processes at industrial scale.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121676321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00041
R. Basiri, M. Popovic, Shehroz S. Khan
Diabetic Foot Ulcer (DFU) is a condition requiring constant monitoring and evaluations for treatment. DFU patient population is on the rise and will soon outpace the available health resources. Autonomous monitoring and evaluation of DFU wounds is a much-needed area in health care. In this paper, we evaluate and identify the most accurate feature extractor that is the core basis for developing a deep learning wound detection network. For the evaluation, we used mAP and F1-score on the publicly available DFU2020 dataset. A combination of UNet and EfficientNetb3 feature extractor resulted in the best evaluation among the 14 networks compared. UNet and Efficientnetb3 can be used as the classifier in the development of a comprehensive DFU domain-specific autonomous wound detection pipeline.
{"title":"Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection","authors":"R. Basiri, M. Popovic, Shehroz S. Khan","doi":"10.1109/ICDMW58026.2022.00041","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00041","url":null,"abstract":"Diabetic Foot Ulcer (DFU) is a condition requiring constant monitoring and evaluations for treatment. DFU patient population is on the rise and will soon outpace the available health resources. Autonomous monitoring and evaluation of DFU wounds is a much-needed area in health care. In this paper, we evaluate and identify the most accurate feature extractor that is the core basis for developing a deep learning wound detection network. For the evaluation, we used mAP and F1-score on the publicly available DFU2020 dataset. A combination of UNet and EfficientNetb3 feature extractor resulted in the best evaluation among the 14 networks compared. UNet and Efficientnetb3 can be used as the classifier in the development of a comprehensive DFU domain-specific autonomous wound detection pipeline.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"52 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120883848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00070
Minqiang Yang, Xinqi Liu, Chengsheng Mao, Bin Hu
Sentiment analysis has become increasingly important in natural language processing (NLP). Recent efforts have been devoted to the graph convolutional network (GCN) due to its advantages in handling the complex information. However, the improvement of GCN in NLP is hindered because the pretrained word vectors do not fit well in various contexts and the traditional edge building methods are not suited well for the long and complex context. To address these problems, we propose the LSTM-GCN model to contextualize the pretrained word vectors and extract the sentiment representations from the complex texts. Particularly, LSTM-GCN captures the sentiment feature representations from multiple different perspectives including context and syntax. In addition to extracting contextual representation from pretrained word vectors, we utilize the dependency parser to analyse the dependency correlation between each word to extract the syntax representation. For each text, we build a graph with each word in the text as a node. Besides the edges between the neighboring words, we also connect the nodes with dependency correlation to capture syntax representations. Moreover, we introduce the message passing mechanism (MPM) which allows the nodes to update their representation by extract information from its neighbors. Also, to improve the message passing performance, we set the edges to be trainable and initialize the edge weights with the pointwise mutual information (PMI) method. The results of the experiments show that our LSTM-GCN model outperforms several state-of-the-art models. And extensive experiments validate the rationality and effectiveness of our model.
{"title":"Graph Convolutional Networks with Dependency Parser towards Multiview Representation Learning for Sentiment Analysis","authors":"Minqiang Yang, Xinqi Liu, Chengsheng Mao, Bin Hu","doi":"10.1109/ICDMW58026.2022.00070","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00070","url":null,"abstract":"Sentiment analysis has become increasingly important in natural language processing (NLP). Recent efforts have been devoted to the graph convolutional network (GCN) due to its advantages in handling the complex information. However, the improvement of GCN in NLP is hindered because the pretrained word vectors do not fit well in various contexts and the traditional edge building methods are not suited well for the long and complex context. To address these problems, we propose the LSTM-GCN model to contextualize the pretrained word vectors and extract the sentiment representations from the complex texts. Particularly, LSTM-GCN captures the sentiment feature representations from multiple different perspectives including context and syntax. In addition to extracting contextual representation from pretrained word vectors, we utilize the dependency parser to analyse the dependency correlation between each word to extract the syntax representation. For each text, we build a graph with each word in the text as a node. Besides the edges between the neighboring words, we also connect the nodes with dependency correlation to capture syntax representations. Moreover, we introduce the message passing mechanism (MPM) which allows the nodes to update their representation by extract information from its neighbors. Also, to improve the message passing performance, we set the edges to be trainable and initialize the edge weights with the pointwise mutual information (PMI) method. The results of the experiments show that our LSTM-GCN model outperforms several state-of-the-art models. And extensive experiments validate the rationality and effectiveness of our model.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121063083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00027
Furkan Gursoy, I. Kakadiaris
As artificial intelligence plays an increasingly substantial role in decisions affecting humans and society, the accountability of automated decision systems has been receiving increasing attention from researchers and practitioners. Fairness, which is concerned with eliminating unjust treatment and discrimination against individuals or sensitive groups, is a critical aspect of accountability. Yet, for evaluating fairness, there is a plethora of fairness metrics in the literature that employ different perspectives and assumptions that are often incompatible. This work focuses on group fairness. Most group fairness metrics desire a parity between selected statistics computed from confusion matrices belonging to different sensitive groups. Generalizing this intuition, this paper proposes a new equal confusion fairness test to check an automated decision system for fairness and a new confusion parity error to quantify the extent of any unfairness. To further analyze the source of potential unfairness, an appropriate post hoc analysis methodology is also presented. The usefulness of the test, metric, and post hoc analysis is demonstrated via a case study on the controversial case of COMPAS, an automated decision system employed in the US to assist judges with assessing recidivism risks. Overall, the methods and metrics provided here may assess automated decision systems' fairness as part of a more extensive accountability assessment, such as those based on the system accountability benchmark.
{"title":"Equal Confusion Fairness: Measuring Group-Based Disparities in Automated Decision Systems","authors":"Furkan Gursoy, I. Kakadiaris","doi":"10.1109/ICDMW58026.2022.00027","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00027","url":null,"abstract":"As artificial intelligence plays an increasingly substantial role in decisions affecting humans and society, the accountability of automated decision systems has been receiving increasing attention from researchers and practitioners. Fairness, which is concerned with eliminating unjust treatment and discrimination against individuals or sensitive groups, is a critical aspect of accountability. Yet, for evaluating fairness, there is a plethora of fairness metrics in the literature that employ different perspectives and assumptions that are often incompatible. This work focuses on group fairness. Most group fairness metrics desire a parity between selected statistics computed from confusion matrices belonging to different sensitive groups. Generalizing this intuition, this paper proposes a new equal confusion fairness test to check an automated decision system for fairness and a new confusion parity error to quantify the extent of any unfairness. To further analyze the source of potential unfairness, an appropriate post hoc analysis methodology is also presented. The usefulness of the test, metric, and post hoc analysis is demonstrated via a case study on the controversial case of COMPAS, an automated decision system employed in the US to assist judges with assessing recidivism risks. Overall, the methods and metrics provided here may assess automated decision systems' fairness as part of a more extensive accountability assessment, such as those based on the system accountability benchmark.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116381773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00021
A. Ramkissoon, Vijayanandh Rajamanickam, W. Goodridge
The existence of fake videos is a problem that is challenging today's social media-enabled world. There are many classifications for fake videos with one of the most popular being DeepFakes. Detecting such fake videos is a challenging issue. This research attempts to comprehend the characteristics that belong to DeepFake videos. In attempting to understand DeepFake videos this work investigates the characteristics of the video that make them unique. As such this research uses scene and texture detection to develop a unique feature set containing 19 data features which is capable of detecting whether a video is a DeepFake or not. This study validates the feature set using a standard dataset of the features relating to the characteristics of the video. These features are analysed using a classification machine learning model. The results of these experiments are examined using four evaluation methodologies. The analysis reveals positive performance with the use of the ML method and the feature set. From these results, it can be ascertained that using the proposed feature set, a video can be predicted as a DeepFake or not and as such prove the hypothesis that there exists a correlation between the characteristics of a video and its genuineness, i.e., whether or not a video is a DeepFake.
{"title":"Scene and Texture Based Feature Set for DeepFake Video Detection","authors":"A. Ramkissoon, Vijayanandh Rajamanickam, W. Goodridge","doi":"10.1109/ICDMW58026.2022.00021","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00021","url":null,"abstract":"The existence of fake videos is a problem that is challenging today's social media-enabled world. There are many classifications for fake videos with one of the most popular being DeepFakes. Detecting such fake videos is a challenging issue. This research attempts to comprehend the characteristics that belong to DeepFake videos. In attempting to understand DeepFake videos this work investigates the characteristics of the video that make them unique. As such this research uses scene and texture detection to develop a unique feature set containing 19 data features which is capable of detecting whether a video is a DeepFake or not. This study validates the feature set using a standard dataset of the features relating to the characteristics of the video. These features are analysed using a classification machine learning model. The results of these experiments are examined using four evaluation methodologies. The analysis reveals positive performance with the use of the ML method and the feature set. From these results, it can be ascertained that using the proposed feature set, a video can be predicted as a DeepFake or not and as such prove the hypothesis that there exists a correlation between the characteristics of a video and its genuineness, i.e., whether or not a video is a DeepFake.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128776656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00049
Fahed Elourajini, Esma Aïmeur
Personality and preferences are essential variables in computational sociology and social science. They describe differences between people at both individual and group levels. In recent years, automated approaches that detect personality traits have received much attention due to the massive availability of individuals' digital footprints. Furthermore, researchers have demonstrated a strong link between personality traits and various downstream tasks such as personalized filtering, profile categorization, and profile embedding. Therefore, the detection of individuals' preferences has become a critical process for improving the performance of different tasks. In this paper, we build on the importance of the individual's behaviour and propose a novel multitask modeling approach that understands and models the users' personalities based on their textual posts and comments within a multimedia framework. The novelties of our work compared to state-of-the-art personality prediction models are: improving the performance of the Big five-factor model (Big5) personality test using shared information from the Myers Briggs Type Indicator (MBTI) test, and proposing a one personality detection framework that accurately predicts both MBTI and Big5 tests simultaneously. Predicting both tests simultaneously improves the personality detection framework's flexibility to be used for different goals instead of being used only for a unique purpose (whether for the MBTI test or for the Big5 test separately). Experiments and results demonstrate that our solution outperforms state-of-the-art models across multiple famous personality datasets.
{"title":"AWS-EP: A Multi-Task Prediction Approach for MBTI/Big5 Personality Tests","authors":"Fahed Elourajini, Esma Aïmeur","doi":"10.1109/ICDMW58026.2022.00049","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00049","url":null,"abstract":"Personality and preferences are essential variables in computational sociology and social science. They describe differences between people at both individual and group levels. In recent years, automated approaches that detect personality traits have received much attention due to the massive availability of individuals' digital footprints. Furthermore, researchers have demonstrated a strong link between personality traits and various downstream tasks such as personalized filtering, profile categorization, and profile embedding. Therefore, the detection of individuals' preferences has become a critical process for improving the performance of different tasks. In this paper, we build on the importance of the individual's behaviour and propose a novel multitask modeling approach that understands and models the users' personalities based on their textual posts and comments within a multimedia framework. The novelties of our work compared to state-of-the-art personality prediction models are: improving the performance of the Big five-factor model (Big5) personality test using shared information from the Myers Briggs Type Indicator (MBTI) test, and proposing a one personality detection framework that accurately predicts both MBTI and Big5 tests simultaneously. Predicting both tests simultaneously improves the personality detection framework's flexibility to be used for different goals instead of being used only for a unique purpose (whether for the MBTI test or for the Big5 test separately). Experiments and results demonstrate that our solution outperforms state-of-the-art models across multiple famous personality datasets.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116998730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00097
Prabin B. Lamichhane, W. Eberle
Many real-world networks, such as computer networks, social networks, and the Internet of Things (loT), can be represented by streaming (or dynamic) graphs. Analysis of these streaming graphs serves as the basis for classification, anomaly detection, community detection, clustering, and visual-ization tasks. This paper uses a Self-Organizing Map (SOM), an unsupervised learning model, to cluster and visualize streaming graphs. As a result, a SOM is used to visualize and interpret the anomaly detection technique on high-dimensional graph-structured data. For this, the SOM-based graph clustering and visualization technique is divided into two phases. In the first phase, we use various existing graph sketching techniques like StreamS pot, SpotLight, and SnapSketch to embed streaming graphs into sketched vectors. Later, in the second phase, we pass the sketched vector inputs into a SOM to cluster and visualize the normal and anomalous graph streams to interpret the anomaly detection technique. In addition, the SOM-based visualization also helps to estimate the quality of embedding (or sketching) techniques.
{"title":"Self-Organizing Map-Based Graph Clustering and Visualization on Streaming Graphs","authors":"Prabin B. Lamichhane, W. Eberle","doi":"10.1109/ICDMW58026.2022.00097","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00097","url":null,"abstract":"Many real-world networks, such as computer networks, social networks, and the Internet of Things (loT), can be represented by streaming (or dynamic) graphs. Analysis of these streaming graphs serves as the basis for classification, anomaly detection, community detection, clustering, and visual-ization tasks. This paper uses a Self-Organizing Map (SOM), an unsupervised learning model, to cluster and visualize streaming graphs. As a result, a SOM is used to visualize and interpret the anomaly detection technique on high-dimensional graph-structured data. For this, the SOM-based graph clustering and visualization technique is divided into two phases. In the first phase, we use various existing graph sketching techniques like StreamS pot, SpotLight, and SnapSketch to embed streaming graphs into sketched vectors. Later, in the second phase, we pass the sketched vector inputs into a SOM to cluster and visualize the normal and anomalous graph streams to interpret the anomaly detection technique. In addition, the SOM-based visualization also helps to estimate the quality of embedding (or sketching) techniques.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131395230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00156
Hao Tian, Shengmin Jin, R. Zafarani
With the demand to model the relationships among three or more entities, higher-order networks are now more widespread across various domains. Relationships such as multiauthor collaborations, co-appearance of keywords, and copurchases can be naturally modeled as higher-order networks. However, due to (1) computational complexity and (2) insufficient higher-order data, exploring higher-order networks is often limited to order-3 motifs (or triangles). To address these problems, we explore and quantify similarites among various network orders. Our goal is to build relationships between different network orders and to solve higher-order problems using lower-order information. Similarities between different orders are not comparable directly. Hence, we introduce a set of general cross-order similarities, and a measure: subedge rate. Our experiments on multiple real-world datasets demonstrate that most higher-order networks have considerable consistency as we move from higher-orders to lower-orders. Utilizing this discovery, we develop a new cross-order framework for higher-order link prediction method. These methods can predict higher-order links from lower-order edges, which cannot be attained by current higher-order methods that rely on data from a single order.
{"title":"Exploiting Cross-Order Patterns and Link Prediction in Higher-Order Networks","authors":"Hao Tian, Shengmin Jin, R. Zafarani","doi":"10.1109/ICDMW58026.2022.00156","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00156","url":null,"abstract":"With the demand to model the relationships among three or more entities, higher-order networks are now more widespread across various domains. Relationships such as multiauthor collaborations, co-appearance of keywords, and copurchases can be naturally modeled as higher-order networks. However, due to (1) computational complexity and (2) insufficient higher-order data, exploring higher-order networks is often limited to order-3 motifs (or triangles). To address these problems, we explore and quantify similarites among various network orders. Our goal is to build relationships between different network orders and to solve higher-order problems using lower-order information. Similarities between different orders are not comparable directly. Hence, we introduce a set of general cross-order similarities, and a measure: subedge rate. Our experiments on multiple real-world datasets demonstrate that most higher-order networks have considerable consistency as we move from higher-orders to lower-orders. Utilizing this discovery, we develop a new cross-order framework for higher-order link prediction method. These methods can predict higher-order links from lower-order edges, which cannot be attained by current higher-order methods that rely on data from a single order.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132817409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00045
Toru Sasaki, Tomonari Masada
Automated Essay Scoring (AES) refers to a set of processes that automatically assigns grades to student-written essays with machine learning models. Existing AES models are mostly trained prompt-specifically with supervised learning, which requires the essay prompt to be accessible to the system vendor at the time of model training. However, essay prompts for high-stakes testing should usually be kept confidential before the test date, which demands the model to be cross-promptly trainable with pre-scored essay data already in hands. Document embeddings obtained from pretrained language models such as Sentence-BERT (sbert) are primarily expected to represent the semantic content of the text. We hypothesize SBERT embeddings also contain assessment-relevant elements that are extractable by document embedding decomposition through Principal Component Analysis (PCA) enhanced with Normalized Discounted Cumulative Gain (nDCG) measurement. The identified evaluative elements in the entire embedding space of the source essays are then cross-promptly transferred to the target essays written on different prompts for binary clustering task of dividing high/low-scored groups. The result implies non-finetuned SBERT already contains evaluative elements to distinguish good and bad essays.
{"title":"Sentence-BERT Distinguishes Good and Bad Essays in Cross-prompt Automated Essay Scoring","authors":"Toru Sasaki, Tomonari Masada","doi":"10.1109/ICDMW58026.2022.00045","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00045","url":null,"abstract":"Automated Essay Scoring (AES) refers to a set of processes that automatically assigns grades to student-written essays with machine learning models. Existing AES models are mostly trained prompt-specifically with supervised learning, which requires the essay prompt to be accessible to the system vendor at the time of model training. However, essay prompts for high-stakes testing should usually be kept confidential before the test date, which demands the model to be cross-promptly trainable with pre-scored essay data already in hands. Document embeddings obtained from pretrained language models such as Sentence-BERT (sbert) are primarily expected to represent the semantic content of the text. We hypothesize SBERT embeddings also contain assessment-relevant elements that are extractable by document embedding decomposition through Principal Component Analysis (PCA) enhanced with Normalized Discounted Cumulative Gain (nDCG) measurement. The identified evaluative elements in the entire embedding space of the source essays are then cross-promptly transferred to the target essays written on different prompts for binary clustering task of dividing high/low-scored groups. The result implies non-finetuned SBERT already contains evaluative elements to distinguish good and bad essays.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128597868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-01DOI: 10.1109/ICDMW58026.2022.00153
Avi Chawla, Nidhi Mulay, M. Bahrami, Vikas Bishnoi, Yatin Katyal, Esteban Moro Egido, Ankur Saraswat, A. Pentland
The COVID-19 pandemic has impacted economic activity not only in the United States, but across the globe. Lockdown and travel restrictions imposed by local authorities have led to change in customer preferences and thus transformation of economic activity from traditional areas to new regions. While most changes have been temporary and short term, some of them have been observed to be of permanent nature. Using large-scale aggregated and anonymized transaction data across various socio-economic groups, we analyse and discuss such temporary relocation of citizens' economic activities in metropolitan areas of 15 states in the US. The results of this study have extensive implications for urban planners and business owners, and can provide insights into the temporary relocation of economic activities resulting from an extreme exogenous shock like the COVID-19 pandemic.
{"title":"Post-pandemic Economic Transformations in the United States of America","authors":"Avi Chawla, Nidhi Mulay, M. Bahrami, Vikas Bishnoi, Yatin Katyal, Esteban Moro Egido, Ankur Saraswat, A. Pentland","doi":"10.1109/ICDMW58026.2022.00153","DOIUrl":"https://doi.org/10.1109/ICDMW58026.2022.00153","url":null,"abstract":"The COVID-19 pandemic has impacted economic activity not only in the United States, but across the globe. Lockdown and travel restrictions imposed by local authorities have led to change in customer preferences and thus transformation of economic activity from traditional areas to new regions. While most changes have been temporary and short term, some of them have been observed to be of permanent nature. Using large-scale aggregated and anonymized transaction data across various socio-economic groups, we analyse and discuss such temporary relocation of citizens' economic activities in metropolitan areas of 15 states in the US. The results of this study have extensive implications for urban planners and business owners, and can provide insights into the temporary relocation of economic activities resulting from an extreme exogenous shock like the COVID-19 pandemic.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134496827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：