Mental health support in colleges is vital in educating students by offering counseling services and organizing supportive events. However, evaluating its effectiveness faces challenges like data collection difficulties and lack of standardized metrics, limiting research scope. Student feedback is crucial for evaluation but often relies on qualitative analysis without systematic investigation using advanced machine learning methods. This paper uses public Student Voice Survey data to analyze student sentiments on mental health support with large language models (LLMs). We created a sentiment analysis dataset, SMILE-College, with human-machine collaboration. The investigation of both traditional machine learning methods and state-of-the-art LLMs showed the best performance of GPT-3.5 and BERT on this new dataset. The analysis highlights challenges in accurately predicting response sentiments and offers practical insights on how LLMs can enhance mental health-related research and improve college mental health services. This data-driven approach will facilitate efficient and informed mental health support evaluation, management, and decision-making.

In recent years, deep learning has been successfully adopted in a wide range of applications related to electronic health records (EHRs) such as representation learning and clinical event prediction. However, due to privacy constraints, limited access to EHR becomes a bottleneck for deep learning research. To mitigate these concerns, generative adversarial networks (GANs) have been successfully used for generating EHR data. However, there are still challenges in high-quality EHR generation, including generating time-series EHR data and imbalanced uncommon diseases. In this work, we propose aMulti-labelTime-seriesGAN(MTGAN) to generate EHR and simultaneously improve the quality of uncommon disease generation. The generator of MTGAN uses a gated recurrent unit (GRU) with a smooth conditional matrix to generate sequences and uncommon diseases. The critic gives scores using Wasserstein distance to recognize real samples from synthetic samples by considering both data and temporal features. We also propose a training strategy to calculate temporal features for real data and stabilize GAN training. Furthermore, we design multiple statistical metrics and prediction tasks to evaluate the generated data. Experimental results demonstrate the quality of the synthetic data and the effectiveness of MTGAN in generating realistic sequential EHR data, especially for uncommon diseases.

Automatic coding of International Classification of Diseases (ICD) is a multi-label text categorization task that involves extracting disease or procedure codes from clinical notes. Despite the application of state-of-the-art natural language processing (NLP) techniques, there are still challenges including limited availability of data due to privacy constraints and the high variability of clinical notes caused by different writing habits of medical professionals and various pathological features of patients. In this work, we investigate the semi-structured nature of clinical notes and propose an automatic algorithm to segment them into sections. To address the variability issues in existing ICD coding models with limited data, we introduce a contrastive pre-training approach on sections using a soft multi-label similarity metric based on tree edit distance. Additionally, we design a masked section training strategy to enable ICD coding models to locate sections related to ICD codes. Extensive experimental results demonstrate that our proposed training strategies effectively enhance the performance of existing ICD coding methods.

Using attention weights to identify information that is important for models’ decision making is a popular approach to interpret attention-based neural networks. This is commonly realized in practice through the generation of a heat-map for every single document based on attention weights. However, this interpretation method is fragile and it is easy to find contradictory examples. In this article, we propose a corpus-level explanation approach, which aims at capturing causal relationships between keywords and model predictions via learning the importance of keywords for predicted labels across a training corpus based on attention weights. Based on this idea, we further propose a concept-based explanation method that can automatically learn higher level concepts and their importance to model prediction tasks. Our concept-based explanation method is built upon a novel Abstraction-Aggregation Network (AAN), which can automatically cluster important keywords during an end-to-end training process. We apply these methods to the document classification task and show that they are powerful in extracting semantically meaningful keywords and concepts. Our consistency analysis results based on an attention-based Naïve Bayes classifier (NBC) also demonstrate that these keywords and concepts are important for model predictions.

Unsupervised aspect detection (UAD) aims at automatically extracting interpretable aspects and identifying aspect-specific segments (such as sentences) from online reviews. However, recent deep learning based topic models, specifically aspect-based autoencoder, suffer from several problems such as extracting noisy aspects and poorly mapping aspects discovered by models to the aspects of interest. To tackle these challenges, in this paper, we first propose a self-supervised contrastive learning framework and an attention-based model equipped with a novel smooth self-attention (SSA) module for the UAD task in order to learn better representations for aspects and review segments. Secondly, we introduce a high-resolution selective mapping (HRSMap) method to efficiently assign aspects discovered by the model to the aspects of interest. We also propose using a knowledge distillation technique to further improve the aspect detection performance. Our methods outperform several recent unsupervised and weakly supervised approaches on publicly available benchmark user review datasets. Aspect interpretation results show that extracted aspects are meaningful, have a good coverage, and can be easily mapped to aspects of interest. Ablation studies and attention weight visualization also demonstrate effectiveness of SSA and the knowledge distillation method.

Representation learning methods for heterogeneous networks produce a low-dimensional vector embedding (that is typically fixed for all tasks) for each node. Many of the existing methods focus on obtaining a static vector representation for a node in a way that is agnostic to the downstream application where it is being used. In practice, however, downstream tasks such as link prediction require specific contextual information that can be extracted from the subgraphs related to the nodes provided as input to the task. To tackle this challenge, we develop , a framework for bridging static representation learning methods using global information from the entire graph with localized attention driven mechanisms to learn contextual node representations. We first pre-train our model in a self-supervised manner by introducing higher-order semantic associations and masking nodes, and then fine-tune our model for a specific link prediction task. Instead of training node representations by aggregating information from all semantic neighbors connected via metapaths, we automatically learn the composition of different metapaths that characterize the context for a specific task without the need for any pre-defined metapaths. significantly outperforms both static and contextual embedding learning methods on several publicly available benchmark network datasets. We also demonstrate the interpretability, effectiveness of contextual learning, and the scalability of through extensive evaluation.

Survival analysis aims at predicting the time to event of interest along with its probability on longitudinal data. It is commonly used to make predictions for a single specific event of interest at a given time point. However, predicting the occurrence of multiple events of interest simultaneously and dynamically is needed in many real-world applications. An intuitive way to solve this problem is to simply apply the standard survival analysis method independently to each prediction task at each time point. However, it often leads to a sub-optimal solution since the underlying dependencies between these tasks are ignored. This motivates us to analyze these prediction tasks jointly in order to select the common features shared across all the tasks. In this paper, we formulate a temporal (Multiple Time points) Multi-Task learning framework (MTMT) for survival analysis problems using tensor representation. More specifically, given a survival dataset and a sequence of time points, which are considered as the monitored time points for the events of interest, we reformulate the survival analysis problem to jointly handle each task at each time point and optimize them simultaneously. We demonstrate the performance of the proposed MTMT model on important real-world datasets, including employee attrition and medical records. We show the superior performance of the MTMT model compared to several state-of-the-art models using standard metrics. We also provide the list of important features selected by our MTMT model thus demonstrating the interpretability of the proposed model.

Electronic medical records (EMR) contain comprehensive patient information and are typically stored in a relational database with multiple tables. Effective and efficient patient information retrieval from EMR data is a challenging task for medical experts. Question-to-SQL generation methods tackle this problem by first predicting the SQL query for a given question about a database, and then, executing the query on the database. However, most of the existing approaches have not been adapted to the healthcare domain due to a lack of healthcare Question-to-SQL dataset for learning models specific to this domain. In addition, wide use of the abbreviation of terminologies and possible typos in questions introduce additional challenges for accurately generating the corresponding SQL queries. In this paper, we tackle these challenges by developing a deep learning based TRanslate-Edit Model for Question-to-SQL (TREQS) generation, which adapts the widely used sequence-to-sequence model to directly generate the SQL query for a given question, and further performs the required edits using an attentive-copying mechanism and task-specific look-up tables. Based on the widely used publicly available electronic medical database, we create a new large-scale Question-SQL pair dataset, named MIMICSQL, in order to perform the Question-to-SQL generation task in healthcare domain. An extensive set of experiments are conducted to evaluate the performance of our proposed model on MIMICSQL. Both quantitative and qualitative experimental results indicate the flexibility and efficiency of our proposed method in predicting condition values and its robustness to random questions with abbreviations and typos.

Integrating regularization methods with standard loss functions such as the least squares, hinge loss, etc., within a regression framework has become a popular choice for researchers to learn predictive models with lower variance and better generalization ability. Regularizers also aid in building interpretable models with high-dimensional data which makes them very appealing. It is observed that each regularizer is uniquely formulated in order to capture data-specific properties such as correlation, structured sparsity and temporal smoothness. The problem of obtaining a consensus among such diverse regularizers while learning a predictive model is extremely important in order to determine the optimal regularizer for the problem. The advantage of such an approach is that it preserves the simplicity of the final model learned by selecting a single candidate model which is not the case with ensemble methods as they use multiple candidate models for prediction. This is called the consensus regularization problem which has not received much attention in the literature due to the inherent difficulty associated with learning and selecting a model from an integrated regularization framework. To solve this problem, in this paper, we propose a method to generate a committee of non-convex regularized linear regression models, and use a consensus criterion to determine the optimal model for prediction. Each corresponding non-convex optimization problem in the committee is solved efficiently using the cyclic-coordinate descent algorithm with the generalized thresholding operator. Our Consensus RegularIzation Selection based Prediction (CRISP) model is evaluated on electronic health records (EHRs) obtained from a large hospital for the congestive heart failure readmission prediction problem. We also evaluate our model on high-dimensional synthetic datasets to assess its performance. The results indicate that CRISP outperforms several state-of-the-art methods such as additive, interactions-based and other competing non-convex regularized linear regression methods.

Predicting event occurrence at the early stage of a longitudinal study is an important and challenging problem which has high practical value in many real-world applications. As opposed to the standard classification and regression problems where a domain expert can provide labels for the data in a reasonably short period of time, training data in such longitudinal studies must be obtained only by waiting for the occurrence of a sufficient number of events. Survival analysis aims at directly predicting the time to an event of interest using the data collected in the past for a certain duration. However, it cannot give an answer to the open question of “how to forecast whether a subject will experience an event by end of a longitudinal study using event occurrence information of other subjects at the early stage of the study?”. The goal of this work is to predict the event occurrence at a future time point using only the information about a limited number of events that occurred at the initial stages of a longitudinal study. This problem exhibits two major challenges: (1) absence of complete information about event occurrence (censoring) and (2) availability of only a partial set of events that occurred during the initial phase of the study. We propose a novel Early Stage Prediction (ESP) framework for building event prediction models which are trained at the early stages of longitudinal studies. First, we develop a novel approach to address the first challenge by introducing a new method for handling censored data using Kaplan-Meier estimator. We then extend the Naive Bayes, Tree-Augmented Naive Bayes (TAN), and Bayesian Network methods based on the proposed framework, and develop three algorithms, namely, ESP-NB, ESP-TAN, and ESPBN, to effectively predict event occurrence using training data obtained at an early stage of the study. More specifically, our approach effectively integrates Bayesian methods with an Accelerated Failure Time (AFT) model by adapting the prior probability of the event occurrence for future time points. The proposed framework is evaluated using a wide range of synthetic and real-world benchmark datasets. Our extensive set of experiments show that the proposed ESP framework is, on an average, 20 percent more accurate compared to existing schemes when using only limited event information in the training data.