With the increased awareness of situations of mental crisis and their societal impact, online services providing emergency support are becoming commonplace in many countries. Computational models, trained on discussions between help-seekers and providers, can support suicide prevention by identifying at-risk individuals. However, the lack of domain-specific models, especially in low-resource languages, poses a significant challenge for the automatic detection of suicide risk. We propose a model that combines pre-trained language models (PLM) with a fixed set of manually crafted (and clinically approved) set of suicidal cues, followed by a two-stage fine-tuning process. Our model achieves 0.91 ROC-AUC and an F2-score of 0.55, significantly outperforming an array of strong baselines even early on in the conversation, which is critical for real-time detection in the field. Moreover, the model performs well across genders and age groups.

IT helpdesks are charged with the task of responding quickly to user queries. To give the user confidence that their query matters, the helpdesk will auto-reply to the user with confirmation that their query has been received and logged. This auto-reply may include generic `boiler-plate' text that addresses common problems of the day, with relevant information and links. The approach explored here is to tailor the content of the auto-reply to the user's problem, so as to increase the relevance of the information included. Problem classification is achieved by training a neural network on a suitable corpus of IT helpdesk email data. While this is no substitute for follow-up by helpdesk agents, the aim is that this system will provide a practical stop-gap.

Social media platforms have become a place where users collaborate, share their ideas and also have conflicts (Hansson et al., 2019; Hansson and Ludwig, 2019). With 126 million active daily users (Shaban, 2019), Twitter is the dominant microblogging platform on which users discuss a breadth of subjects and even play a role in influencing current trends. Users on Twitter post short and often informal messages (tweets) in which they share information and project opinions and sentiments about what is going on in the world. Twitter has been a major platform for sharing scholarly articles, and many researchers have used it to develop various metrics for scholarly articles (Haustein, 2019). Other social media platforms like Facebook and Weibo have also been sources to study online users' responses (Kou et al., 2017). Social media platforms have become a hub where users express their opinions and emotions related to multiple fields of interest (Chatterjee et al., 2019). Researchers have studied the sentiments and emotions associated with research articles on these platforms (Freeman et al., 2019, 2020).

This paper analyses the degree to which dialect classifiers based on syntactic representations remain stable over space and time. While previous work has shown that the combination of grammar induction and geospatial text classification produces robust dialect models, we do not know what influence both changing grammars and changing populations have on dialect models. This paper constructs a test set for 12 dialects of English that spans three years at monthly intervals with a fixed spatial distribution across 1,120 cities. Syntactic representations are formulated within the usage-based Construction Grammar paradigm (CxG). The decay rate of classification performance for each dialect over time allows us to identify regions undergoing syntactic change. And the distribution of classification accuracy within dialect regions allows us to identify the degree to which the grammar of a dialect is internally heterogeneous. The main contribution of this paper is to show that a rigorous evaluation of dialect classification models can be used to find both variation over space and change over time.

Healthcare is an important aspect of human life. Use of technologies in healthcare has increased manifolds after the pandemic. Internet of Things based systems and devices proposed in literature can help elders, children and adults facing/experiencing health problems. This paper exhaustively reviews thirty-nine wearable based datasets which can be used for evaluating the system to recognize Activities of Daily Living and Falls. A comparative analysis on the SisFall dataset using five machine learning methods i.e., Logistic Regression, Linear Discriminant Analysis, K-Nearest Neighbor, Decision Tree and Naive Bayes is performed in python. The dataset is modified in two ways, in first all the attributes present in dataset are used as it is and labelled in binary form. In second, magnitude of three axes(x,y,z) for three sensors value are computed and then used in experiment with label attribute. The experiments are performed on one subject, ten subjects and all the subjects and compared in terms of accuracy, precision and recall. The results obtained from this study proves that KNN outperforms other machine learning methods in terms of accuracy, precision and recall. It is also concluded that personalization of data improves accuracy.

Artificial intelligence (AI) models have been developed to predict clinically relevant biomarkers for colorectal cancer (CRC), including microsatellite instability (MSI). However, existing deep-learning networks are data-hungry and require large training datasets, which are often lacking in the medical domain. In this study, based on the latest Hierarchical Vision Transformer using Shifted Windows (Swin-T), we developed an efficient workflow for biomarkers in CRC (MSI, hypermutation, chromosomal instability, CpG island methylator phenotype, BRAF, and TP53 mutation) that required relatively small datasets, but achieved a state-of-the-art (SOTA) predictive performance. Our Swin-T workflow substantially outperformed published models in an intra-study cross-validation experiment using the TCGA-CRC-DX dataset (N = 462). It also demonstrated excellent generalizability in cross-study external validation and delivered a SOTA AUROC of 0.90 for MSI, using the MCO dataset for training (N = 1065) and the TCGA-CRC-DX for testing. A similar performance (AUROC = 0.91) was achieved by Echle et al., using ~8000 training samples (ResNet18) on the same testing dataset. Swin-T was extremely efficient when using small training datasets and exhibited robust predictive performance with 200-500 training samples. These data indicate that Swin-T could be 5-10 times more efficient than existing algorithms for MSI based on ResNet18 and ShuffleNet. Furthermore, the Swin-T models showed promise as pre-screening tests for MSI status and BRAF mutation status, which could exclude and reduce the samples before subsequent standard testing in a cascading diagnostic workflow, to allow a reduction in turnaround time and costs.

Dementia is a neuropsychiatric brain disorder that usually occurs when one or more brain cells stop working partially or at all. Diagnosis of this disorder in the early phases of the disease is a vital task to rescue patients lives from bad consequences and provide them with better healthcare. Machine learning methods have been proven to be accurate in predicting dementia in the early phases of the disease. The prediction of dementia depends heavily on the type of collected data which usually are gathered from Normalized Whole Brain Volume (nWBV) and Atlas Scaling Factor (ASF) which are normally measured and corrected from Magnetic Resonance Imaging (MRIs). Other biological features such as age and gender can also help in the diagnosis of dementia. Although many studies use machine learning for predicting dementia, we could not reach a conclusion on the stability of these methods for which one is more accurate under different experimental conditions. Therefore, this paper investigates the conclusion stability regarding the performance of machine learning algorithms for dementia prediction. To accomplish this, a large number of experiments were run using 7 machine learning algorithms and two feature reduction algorithms namely, Information Gain (IG) and Principal Component Analysis (PCA). To examine the stability of these algorithms, thresholds of feature selection were changed for the IG from 20% to 100% and the PCA dimension from 2 to 8. This has resulted in 7x9 + 7x7= 112 experiments. In each experiment, various classification evaluation data were recorded. The obtained results show that among seven algorithms the support vector machine and Naive Bayes are the most stable algorithms while changing the selection threshold. Also, it was found that using IG would seem more efficient than using PCA for predicting Dementia.

Cough audio signal classification is a potentially useful tool in screening for respiratory disorders, such as COVID-19. Since it is dangerous to collect data from patients with such contagious diseases, many research teams have turned to crowdsourcing to quickly gather cough sound data, as it was done to generate the COUGHVID dataset. The COUGHVID dataset enlisted expert physicians to diagnose the underlying diseases present in a limited number of uploaded recordings. However, this approach suffers from potential mislabeling of the coughs, as well as notable disagreement between experts. In this work, we use a semi-supervised learning (SSL) approach to improve the labeling consistency of the COUGHVID dataset and the robustness of COVID-19 versus healthy cough sound classification. First, we leverage existing SSL expert knowledge aggregation techniques to overcome the labeling inconsistencies and sparsity in the dataset. Next, our SSL approach is used to identify a subsample of re-labeled COUGHVID audio samples that can be used to train or augment future cough classification models. The consistency of the re-labeled data is demonstrated in that it exhibits a high degree of class separability, 3x higher than that of the user-labeled data, despite the expert label inconsistency present in the original dataset. Furthermore, the spectral differences in the user-labeled audio segments are amplified in the re-labeled data, resulting in significantly different power spectral densities between healthy and COVID-19 coughs, which demonstrates both the increased consistency of the new dataset and its explainability from an acoustic perspective. Finally, we demonstrate how the re-labeled dataset can be used to train a cough classifier. This SSL approach can be used to combine the medical knowledge of several experts to improve the database consistency for any diagnostic classification task.

graph neural networks (GNNs) are the dominant paradigm for modeling and handling graph structure data by learning universal node representation. The traditional way of training GNNs depends on a great many labeled data, which results in high requirements on cost and time. In some special scene, it is even unavailable and impracticable. Self-supervised representation learning, which can generate labels by graph structure data itself, is a potential approach to tackle this problem. And turning to research on self-supervised learning problem for heterogeneous graphs is more challenging than dealing with homogeneous graphs, also there are fewer studies about it. In this paper, we propose a SElfsupervised learning method for heterogeneous graph via Structure Information based on Metapath (SESIM). The proposed model can construct pretext tasks by predicting jump number between nodes in each metapath to improve the representation ability of primary task. In order to predict jump number, SESIM uses data itself to generate labels, avoiding time-consuming manual labeling. Moreover, predicting jump number in each metapath can effectively utilize graph structure information, which is the essential property between nodes. Therefore, SESIM deepens the understanding of models for graph structure. At last, we train primary task and pretext tasks jointly, and use meta-learning to balance the contribution of pretext tasks for primary task. Empirical results validate the performance of SESIM method and demonstrate that this method can improve the representation ability of traditional neural networks on link prediction task and node classification task.

We present the first dataset and evaluation results on a newly defined computational task of trigger warning assignment. Labeled corpus data has been compiled from narrative works hosted on Archive of Our Own (AO3), a well-known fanfiction site. In this paper, we focus on the most frequently assigned trigger type--violence--and define a document-level binary classification task of whether or not to assign a violence trigger warning to a fanfiction, exploiting warning labels provided by AO3 authors. SVM and BERT models trained in four evaluation setups on the corpora we compiled yield $F_1$ results ranging from 0.585 to 0.798, proving the violence trigger warning assignment to be a doable, however, non-trivial task.