A unified hierarchical attention framework for sequential recommendation by fusing long and short-term preferences

To classify the image material on the internet, the deep learning methodology, especially deep neural network, is the most optimal and costliest method of all computer vision methods. Convolutional neural networks (CNNs) learn a comprehensive feature representation by exploiting local information with a fixed receptive field, demonstrating distinguished capacities on image classification. Recent works concentrate on efficient feature exploration, which neglect the global information for holistic consideration. There is large effort to reduce the computational costs of deep neural networks. Here, we provide a hierarchical global attention mechanism that improve the network representation with restricted increase of computation complexity. Different from nonlocal-based methods, the hierarchical global attention mechanism requires no matrix multiplication and can be flexibly applied in various modern network designs. Experimental results demonstrate that proposed hierarchical global attention mechanism can conspicuously improve the image classification precision—a reduction of 7.94% and 16.63% percent in Top 1 and Top 5 errors separately—with little increase of computation complexity (6.23%) in comparison to competing approaches.

Recent years have witnessed the rapidly growing of the amount of semistructured documents in real-world applications. Due to the huge size of the real-world data, how to manage semistructured documents effectively is a big challenge for researchers. As a fundamental task in natural language processing field, document classification is a feasible way to handle the large-scale semistructured documents. However, existing methods fail to explicitly take advantage of the hierarchical semantics in semistructured documents. It's known that the contained semantics is beneficial for understanding the semistructured documents. Considering the hierarchical structure of a given semistructured document, we propose a semistructured document classification framework which explicitly utilizes the semantic hierarchical attention mechanism. More specifically, the hierarchical attention mechanism and graph neural network are employed to model semistructured documents, by which the multilevel semantic relationships and grammatical information are considered. Moreover, we propose an adaptive class cost learning method to treat the issue of data imbalance. Comprehensive experiments are conducted on two real-world data sets, and the results demonstrate that our framework performs better than selected baselines for semistructured document classification.

The automated processing of Electronic Health Records (EHRs) poses a significant challenge due to their unstructured nature, rich in valuable, yet disorganized information. Natural Language Processing (NLP), particularly Named Entity Recognition (NER), has been instrumental in extracting structured information from EHR data. However, existing literature primarly focuses on extracting handcrafted clinical features through NLP and NER methods without delving into their learned representations. In this work, we explore the untapped potential of these representations by considering their contextual richness and entity-specific information. Our proposed methodology extracts representations generated by a transformer-based NER model on EHRs data, combines them using a hierarchical attention mechanism, and employs the obtained enriched representation as input for a clinical prediction model. Specifically, this study addresses Overall Survival (OS) in Non-Small Cell Lung Cancer (NSCLC) using unstructured EHRs data collected from an Italian clinical centre encompassing 838 records from 231 lung cancer patients. Whilst our study is applied on EHRs written in Italian, it serves as use case to prove the effectiveness of extracting and employing high level textual representations that capture relevant information as named entities. Our methodology is interpretable because the hierarchical attention mechanism highlights the information in EHRs that the model considers the most crucial during the decision-making process. We validated this interpretability by measuring the agreement of domain experts on the importance assigned by the hierarchical attention mechanism to EHRs information through a questionnaire. Results demonstrate the effectiveness of our method, showcasing statistically significant improvements over traditional manually extracted clinical features.

With an increase in online longitudinal users’ interactions, capturing users’ precise preferences and giving accurate recommendations have become an urgent need for all businesses. Existing sequence-aware methods generally exploit a static low-rank vector for acquiring the overall sequential features, and incorporate context information as auxiliary input. As a result, they have a restricted modeling ability for extracting multi-grained sequential behaviors over contextual information. In other words, they poorly capture the hierarchical relationship between context relations and item relations that currently influence users’ preferences in a unified framework. Besides, they usually utilize users’ short-term preferences with either static or irrelevant long-term representation for the prediction. To tackle the above issues, in this paper, we propose a novel Context-aware Recommender System Based on a Deep Sequential Learning Approach (CReS) to capture users’ dynamic preferences by modeling the hierarchical relationships between contexts and items in a particular session, and for combining users’ short-term sessions with the relevant long-term representations. Specifically, within a certain session, we design a hierarchical attention network between the identified context relations and items relations, namely CReSession. Therefore, with CReSession, we could provide a suitable session representation that mimics the hierarchical user interests on multiple granularities of contextual types and its corresponding items. We then introduce a neural attentive bi-directional GRU network to distill only those highly related to the recent short-term session. Finally, the relevant long-term representations and the short-term session are combined with the sequential residual connection to form the final user representation in a unified manner. With extensive experiments on two real-world datasets, CReS not only achieves significant improvement over the state-of-the-art methods in terms of pre-defined metrics, but also provides an interpretable result regarding why we recommend these items to users.

Recognizing text in images has been a hot research topic in computer vision for decades due to its various application. However, the variations in text appearance in term of perspective distortion, text line curvature, text styles, etc., cause great trouble in text recognition. Inspired by the Transformer structure [1] that achieved outstanding performance in many natural language processing related applications, we propose a new Transformer-like structure for text recognition in images, which is referred to as the Hierarchical Attention Transformer Network (HATN). The entire network can be trained end-to-end by using only images and sentence-level annotations. A new hierarchical attention mechanism is proposed to lean the character-level, word-level and sentence-level contexts more efficiently and sufficiently. Extensive experiments on seven public datasets with regular and irregular text arrangements have demonstrated that the proposed HATN can achieve accurate recognition results with high efficiency.

This paper presents a new architecture for multimodal sentiment analysis exploiting hierarchical cross-modal attention mechanisms, as well as two parallel lanes for audio analysis. Traditional sentiment analysis approaches are mainly based on text data, which can be inefficient as valuable sentiment information may reside within images and audio. Aiming at solving this issue, the model provides a unified framework that integrates three modalities (text, image, audio) based on BERT text encoder, ResNet50 visual features extractor and hybrid CNN-Wav2Vec2.0 pipeline for audio representation. Specifically, its main innovation is a dual audio pathway augmented with a dynamic gating module and a cross-modal self-attention layer that enables fine-grained interaction among modalities. Our model reports state-of-the-art performance on various benchmarks, outperforming recent approaches: CLIP, MISA and MSFNet. Such that, the results reveal an improvement of classification accuracy especially with missing or noisy modality data. The system has robustness and reliability, which is validated with an exhaustive analysis through metrics like precision, recall, F1-score, and confusion-matrices. In addition, such an architecture demonstrates modular scalability and adaptability across domains, making it proficient for applications in healthcare, social media, and customer service. By providing a framework for developing affective AI systems that can decode human emotion from intricate multimodal features, the study lays the groundwork for future research into further processing of such data streams in the longer term, including real-time processing, domain-specific adjustments, and extending the analysis to the addition of multi-channel sensor input combining physiological and temporal data streams.

Sequence labeling is a basic task in natural language processing, which is of great help to processing text information. Conventional sequence labeling approaches heavily rely on hand-crafted or language-specific features, which requires a lot of time. Therefore, most of the existing methods are based on the BiLSTM-CRF model, but how to use a neural network to extract useful information for each unit or segment in the input sequence becomes the main factor limiting the efficiency. Several BiLSTM-CRF based models for sequence labeling have been presented, but the major limitation is how to use neural networks for extracting useful representations for each unit or segment in the input sequence. In response to this problem, this paper proposes a sequence labeling algorithm based on hierarchical features and attention mechanism, which uses a hierarchical structure to integrate character-level and word-level information, and applies different attention mechanisms to these two layers of information. According to the structural characteristics of different levels, excavate more potential information. Finally, the previously captured and guided features are used for sequence tag prediction using CRF. Finally, the proposed model is subjected to comparative experiments and the results obtained are analyzed.

Purpose Music therapy serves as an effective non-pharmacological intervention in rehabilitation, promoting cognitive, emotional, and motor recovery. This study aims to enhance the precision and adaptability of emotion-aware music therapy by developing a hierarchical attention-based deep learning framework that aligns therapeutic music interventions with patients’ dynamic emotional and contextual states. Materials and Methods A hierarchical attention mechanism was integrated into a deep learning architecture to process multi-level contextual cues, including physiological signals, behavioural responses, and environmental parameters. The model was trained to generate a context-aware emotion mapping that captures both short-term affective fluctuations and long-term emotional patterns. The hierarchical attention layers were designed to prioritise salient features across temporal and contextual dimensions, enabling adaptive selection of therapeutic music stimuli. Results Preliminary experimental evaluations demonstrated that the proposed framework improves personalisation in music selection, enhances emotional regulation, and increases patient engagement during rehabilitation sessions. The system effectively adapted to individual emotional trajectories, offering a more responsive and patient-centred therapeutic experience compared to conventional static music therapy methods. Conclusions The hierarchical attention-based emotion-aware model shows strong potential in advancing intelligent music therapy systems. By bridging neuroscience, artificial intelligence, and therapeutic practice, this approach contributes to more effective and personalised rehabilitation outcomes through context-aware, adaptive music interventions.

Next item recommendation is an important yet challenging task in real-world applications such as E-commerce. Since people often carry out a series of online shopping activities, in order to predict what a user may purchase next, it is essential to model the user's general taste as well as the sequential correlation between purchases. Existing models combine these two factors directly without considering the dynamic changes of a user's long-term and short-term preferences. Meanwhile, when a purchase session contains multiple items, not all of them have the same impact on the next item to purchase. To address these limitations, we propose a model that introduces hierarchical attention to dynamically balance between general taste (long-term preference) and sequential behavior (short-term preference). To weight individual items in the same session, we design a neural memory network with attention mechanism to learn the dynamic weights. Our model can adapt the embedding of each session as well as the embedding of long-term and short-term preferences. Extensive experiments on three real-world datasets show that our model significantly outperforms state-of-the-art methods based on commonly used evaluation metrics.

Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational costs. To overcome these issues, we introduce the Channel-Aware U-Shaped Mamba (CU-Mamba) model, which incorporates a dual State Space Model (SSM) framework into the U-Net architecture. Build on top of hierarchical structure and attention mechanism, CU-Mamba employs a Spatial SSM module for global context encoding and a Channel SSM component to preserve channel correlation features, both in linear computational complexity relative to the feature map size. Extensive experimental results validate CU-Mamba's superiority over existing state-of-the-art methods, underscoring the importance of integrating both spatial and channel contexts in image restoration.

A hierarchical attention detector for bearing surface defect detection

Detecting bias in text is an increasingly relevant topic, given the information overload problem. Automating this task is crucial for our needs of quality news consumption. With this in mind, we explore modern deep learning approaches, including contextualized word embeddings and attention mechanisms, to compare the effects of different document representation choices. We design token-wise, sentence-wise and hierarchical document representations. Focusing on hyperpartisan news detection, we show that hierarchical attention mechanisms are able to better capture information at different levels of granularity (including intra and inter-sentence), which seems to be relevant for this task. With an accuracy of 82.5%, our best performing system is based on an ensemble of hierarchical attention networks with ELMo embeddings, achieving state-of-the-art performance on the SemEval-2019 Task4 dataset.

HiAM: A Hierarchical Attention based Model for knowledge graph multi-hop reasoning

Integration of short and long-term interests: A preference aware session-based recommender

An increased interest in sequential recommendation has been observed in recent years. Many models have been proposed to leverage the sequential user-item interaction data, which includes those based on Markov Chain or recurrent neural networks. Most of these models are designed for the scenario where each historical record composed of single item. However, the records could be a subset of items (or session) such as music playlists and baskets in e-commerce applications. How to leverage the session structure to improve the effectiveness of the recommendation system is a challenge. To this end, we propose a MEmory-augmented Attention Network for Sequential recommendation (MEANS), to effectively recommend next items given the sequential session data. The most recent sessions are stored into external memory after a max-pooling operation. The long-term user preference are learned through an attention network which is stacked on the memory layer. Finally, the mixture of long-term and short-term preference is feeded into the prediction layer to make recommendations. Extensive experiments on four real datasets show that MEANS outperforms various state-of-the-art sequential recommendation models.