Sequential Pattern Research Articles

Abstract WP109: Framework for Evaluating Sequential Patterns in Post-Acute Transitions of Care Among Ischemic and Hemorrhagic Stroke Survivors: Analysis of Medicare Beneficiaries in the State of Texas

Background: Utilizing medical claims derived information, we evaluated temporal trends in post-acute care utilization pathways among patients with acute ischemic stroke (AIS) or intracerebral hemorrhage (ICH). Methods: Data were retrieved from a CMS Qualified Entity housing healthcare utilization data for ≥80% of the Texas state population (100% of Medicare Fee-for-Service). Analytical sample included all Medicare enrollees with a primary discharge diagnosis (AIS or ICH) from 2016 to 2020. Episodes of care within 1 year of discharge were collated and categorized “Discharged”, “Inpatient Rehab”, “Skilled Nursing”, “Hospice”, “Readmission”, or “Death”. Sequential pattern mining was used to calculate conditional probabilities (likelihoods) of post-discharge transitions of care. Prediction rules were generated by converting care sequences into antecedent-consequent pairs. Rule based performance was evaluated by extracting 90 day sequences (antecedent) and assigning predictions of 1 year settings (consequent) based on highest likelihoods (Panel A). Results: A total of 82,829 AIS and 12,761 ICH hospitalizations were analyzed (described in Panel B). An example rulebook of care sequences and associated likelihoods of subsequent transitions for ICH is depicted in Panel C. Whereas ICH survivors entering inpatient rehab had respective likelihoods of 11.2% and 16.2% for subsequently being admitted to hospice care or death, those entering skilled nursing facilities had respective likelihoods of 21.9% and 30.5%. Using 90 day transition of care sequences, predictive accuracy of exact 1 year setting for AIS was 65.3% (1 attempt) and 84.0% (2 attempts); for ICH, the respective accuracy was 50.9% and 82.3% (Panel D). Conclusion: Post stroke transitions of care patterns may yield insights as to the long-term trajectory of outcomes. Predictive capability is expected to strengthen with planned multi-modal integration of EMR-derived clinical and imaging features.

Using Natural Language Processing and Patient Journey Clustering for Temporal Phenotyping of Antimicrobial Therapies for Cat Bite Abscesses

BackgroundTemporal phenotyping of patient journeys, which capture the common sequence patterns of interventions in the treatment of a specific condition, is useful to support understanding of antimicrobial usage in veterinary patients. Identifying and describing these phenotypes can inform antimicrobial stewardship programs designed to fight antimicrobial resistance, a major health crisis affecting both humans and animals, in which veterinarians have an important role to play. ObjectiveThis research proposes a framework for extracting temporal phenotypes of patient journeys from clinical practice data through the application of natural language processing (NLP) and unsupervised machine learning (ML) techniques, using cat bite abscesses as a model condition. By constructing temporal phenotypes from key events, the relationship between antimicrobial administration and surgical interventions can be described, and similar treatment patterns can be grouped together to describe outcomes associated with specific antimicrobial selection. MethodsCases identified as having a cat bite abscess as a diagnosis were extracted from VetCompass Australia, a database of veterinary clinical records. A classifier was trained and used to label the most clinically relevant event features in each record as chosen by a group of veterinarians. The labeled records were processed into coded character strings, where each letter represents a summary of specific types of treatments performed at a given visit. The sequences of letters representing the cases were clustered based on weighted Levenshtein edit distances with KMeans++ to identify the main variations of the patient treatment journeys, including the antimicrobials used and their duration of administration. ResultsA total of 13,744 records that met the selection criteria was extracted and grouped into 8,436 cases. There were 9 clinically distinct event sequence patterns (temporal phenotypes) of patient journeys identified, representing the main sequences in which surgery and antimicrobial interventions are performed. Patients receiving amoxicillin and surgery had the shortest duration of antimicrobial administration (median of 3.4 days) and patients receiving cefovecin with no surgical intervention had the longest antimicrobial treatment duration (median of 27 days). ConclusionOur study demonstrates methods to extract and provide an overview of temporal phenotypes of patient journeys, which can be applied to text-based clinical records for multiple species or clinical conditions. We demonstrate the effectiveness of this approach to derive real-world evidence of treatment impacts using cat bite abscesses as a model condition to describe patterns of antimicrobial therapy prescriptions and their outcomes.

Lipid membrane damage by amyloid peptides: Effect of the polar/non-polar sequence pattern

Lipid membrane damage by amyloid peptides: Effect of the polar/non-polar sequence pattern

Effect of polar/non-polar sequence pattern on peptide aggregation in hydrophobic and hydrophilic solvents

Effect of polar/non-polar sequence pattern on peptide aggregation in hydrophobic and hydrophilic solvents

Real-world sequential treatment patterns and clinical outcomes among patients with advanced urothelial carcinoma in Japan.

Immune checkpoint inhibitors and enfortumab vedotin have opened new avenues for sequential treatment strategies for locally advanced/metastatic urothelial carcinoma (la/mUC). In the pre-enfortumab vedotin era, many patients could not receive third-line treatment owing to rapid disease progression and poor general status. This study aimed to analyze real-world sequential treatment practices for la/mUC in Japan, with a focus on patients who do not receive third-line treatment. We analyzed data for 1023 la/mUC patients diagnosed between January 2020 and December 2021 at 54 institutions from a Japanese nationwide cohort. At the median follow-up of 28.5 months, the median overall survival from first-line initiation for 905 patients who received systemic anticancer treatment was 19.1 months. Among them, 81% and 32% received second- and third-line treatment. Notably, 52% had their treatment terminated before the opportunity for third-line treatment. Multivariate logistic regression analysis revealed that low performance status (≥1), elevated neutrophil-to-lymphocyte ratio (≥3), and low body mass index (<21 kg/m2) at the start of first-line treatment were independent risk factors for not proceeding to third-line treatment (p = 0.0024, 0.0069, and 0.0058, respectively). In this cohort, 33% had one of these factors, 36% had two, and 15% had all three. This study highlights the high frequency of factors associated with poor tolerance to anticancer treatment in la/mUC patients. The findings suggest the need to establish optimal sequential treatment strategies, maximizing efficacy within time and tolerance constraints, while concurrently providing strong supportive care, considering immunological and nutritional aspects.

Application-Oriented Pattern Analysis Utilizing the SPTA Algorithm for Efficient Targeted Sequential Pattern Mining

In the gaming business, esports, also known as competitive gaming, has ushered in a new age that has brought up some innovative issues. One of these challenges is determining the level of player competency based on their plans and talents. Using game traces from Rocket League, a one-of-a-kind "soccer with rocket-powered cars" game, this investigation focuses on the automatic recognition of skill shots, an essential component of player performance. Challenges arise for classic pattern matching algorithms due to the unique characteristics of each skill performance. A Player Skill Modeling through Data-Centric Patterns (PSM-DCP) strategy that uses pattern mining and supervised learning techniques offers a solution to this issue. To highlight the potential of our system for player modeling shows that it can efficiently detect a wide variety of Rocket League abilities and shots. Also confirm the efficacy of this technique by conducting a comprehensive set of tests, which reveals that it is able to effectively differentiate and evaluate skillshots. The study results have a wide range of implications for various applications within esports. These applications include the improvement of match-making algorithms, the provision of assistance to game commentators in the form of insightful assessments, and the development of learning systems that are intended to improve player abilities. This study not only tackles the unique context of Rocket League, but it also offers a framework for exploiting data-centric techniques in esports analytics. This allows for additional developments and applications in competitive gaming, which opens up new paths for possibilities.

Deep sequential pattern mining for readability enhancement of Indonesian summarization

In text summarization research, readability is a great issue that must be addressed. Our hypothesis is readability can be accomplished by using text representations that keep the meaning of text documents intact. Therefore, this study aims to combine sequential pattern mining (SPM) in producing a sequence of a word as text representation with unsupervised deep learning to produce an Indonesian text summary called DeepSPM. This research uses PrefixSpan as an SPM algorithm and deep belief network (DBN) as an unsupervised deep learning method. This research uses 18,774 Indonesian news text from IndoSum. The readability aspect is evaluated by recall-oriented understudy for gisting evaluation (ROUGE) as a co-selection-based analysis; Dwiyanto Djoko Pranowo metrics, Gunning fog index (GFI), and Flesch-Kincaid grade level (FKGL) as content-based analysis; and human readability evaluation with two experts. The experiment result shows that DeepSPM yields better than DBN, with the F-measure value of ROUGE-1 enhanced to 0.462, ROUGE-2 is 0.37, and ROUGE-L is 0.41. The significance of ROUGE results also be tested using T-Test. The content-based analysis and human readability evaluation findings are conformable with the findings of co-selection-based analysis that generated summaries are only partially readable or have a medium level of readability aspect.

An effective deep learning-based approach for splice site identification in gene expression.

A crucial stage in eukaryote gene expression involves mRNA splicing by a protein assembly known as the spliceosome. This step significantly contributes to generating and properly operating the ultimate gene product. Since non-coding introns disrupt eukaryotic genes, splicing entails the elimination of introns and joining exons to create a functional mRNA molecule. Nevertheless, accurately finding splice sequence sites using various molecular biology techniques and other biological approaches is complex and time-consuming. This paper presents a precise and reliable computer-aided diagnosis (CAD) technique for the rapid and correct identification of splice site sequences. The proposed deep learning-based framework uses long short-term memory (LSTM) to extract distinct patterns from RNA sequences, enabling rapid and accurate point mutation sequence mapping. The proposed network employs one-hot encodings to find sequential patterns that effectively identify splicing sites. A thorough ablation study of traditional machine learning, one-dimensional convolutional neural networks (1D-CNNs), and recurrent neural networks (RNNs) models was conducted. The proposed LSTM network outperformed existing state-of-the-art approaches, improving accuracy by 3% and 2% for the acceptor and donor sites datasets.

Using methylation data to improve transcription factor binding prediction

ABSTRACT Modelling the regulatory mechanisms that determine cell fate, response to external perturbation, and disease state depends on measuring many factors, a task made more difficult by the plasticity of the epigenome. Scanning the genome for the sequence patterns defined by Position Weight Matrices (PWM) can be used to estimate transcription factor (TF) binding locations. However, this approach does not incorporate information regarding the epigenetic context necessary for TF binding. CpG methylation is an epigenetic mark influenced by environmental factors that is commonly assayed in human cohort studies. We developed a framework to score inferred TF binding locations using methylation data. We intersected motif locations identified using PWMs with methylation information captured in both whole-genome bisulfite sequencing and Illumina EPIC array data for six cell lines, scored motif locations based on these data, and compared with experimental data characterizing TF binding (ChIP-seq). We found that for most TFs, binding prediction improves using methylation-based scoring compared to standard PWM-scores. We also illustrate that our approach can be generalized to infer TF binding when methylation information is only proximally available, i.e. measured for nearby CpGs that do not directly overlap with a motif location. Overall, our approach provides a framework for inferring context-specific TF binding using methylation data. Importantly, the availability of DNA methylation data in existing patient populations provides an opportunity to use our approach to understand the impact of methylation on gene regulatory processes in the context of human disease.

Effects of hydrophobic sequence patterns on disordered protein conformational ensembles

Effects of hydrophobic sequence patterns on disordered protein conformational ensembles

SN-RNSP: Mining self-adaptive nonoverlapping repetitive negative sequential patterns in transaction sequences

Negative sequential patterns (NSP) focus on non-occurring events and play a role that cannot be replaced by positive sequential patterns (PSP). Considering the repetitive occurrence of sequential patterns in a sequence, repetitive NSP (RNSP) mining captures frequent NSP across different sequences from a database. Those patterns benefit many tasks of transaction services, e.g., fraud detection and medical diagnosis. However, limited studies focusing on mining RNSP are proposed, e.g., e-RNSP and ONP-Miner, and they are devised under strict constraints and are inefficient in practice. To address these issues, this paper proposes a Self-adaptive Nonoverlapping RNSP mining method SN-RNSP to mine nonoverlapping RNSP with the self-adaptive gap between successive elements from transaction sequences, which requires that each element cannot be reused at the same position in occurrences, and the gap value does not need to be specified in advance. First, this paper develops a method that maintains occurrences of pattern candidates via the bitmap structure to capture all repetitive PSP (RPSP), which utilizes the bitmap-based operation to calculate support efficiently. Second, SN-RNSP leverages bitmaps to record the locations of RPSP and RNSP in the database and query the repetition times of corresponding RPSP for the support calculation of RNSP. Conducted on real-world and synthetic datasets, extensive experiments demonstrate that SN-RNSP can discover more patterns with better mining performance than the state-of-the-art RNSP mining algorithms in transaction sequence databases.

Comprehensive analysis to evaluate the molecular biological mechanism and clinical efficacy of Jian Pi Yang Zheng Xiao Zheng in the treatment of gastric cancer

BackgroundClinically, Jian Pi Yang Zheng Xiao Zheng (JPYZXZ) has been used for many years to treat gastric cancer (GC); however, the underlying mechanism and targets are still undetermined. MethodsThe active JPYZXZ components were assessed by Liquid chromatography combined with quadrupole-orbitrap high-resolution mass spectrometry (LC-Q-orbitrap-HRMS). These findings were combined with the TCM Systematic Pharmacology Database and Analysis Platform (TCMSP) database to perform network pharmacological analysis. Using second-generation sequencing on nude mouse samples and single-cell sequencing on human GC tissue, we investigated the potential targets and mechanisms of action of JPYZXZ. A multi-center clinical trial (www.chictr.org.cn ChiCTR1900028147) validated our findings. ResultsA total of 869 compounds were detected using LC-Q-orbitrap-HRMS analysis of JPYZXZ granules in order to examine the underlying mechanism of JPYZXZ in the treatment of GC. The network pharmacology analysis indicated that the potential mechanism of JPYZXZ in treating GC is associated with multiple molecular mechanisms and signaling pathways. The overlap between target genes was identified by second-generation sequencing and gene expression patterns obtained from single-cell transcriptome sequencing. We identified many crucial genes, including DZIP1, FSTL3, USP51, LHFPL6, and NREP. Subsequent investigations have consistently demonstrated that the upregulation of these genes is strongly associated with the development of an immunosuppressive microenvironment. JPYZXZ has been shown in clinical trials to reduce the number of immunosuppressive cell types in patients' peripheral blood, lower the levels of inflammatory factors, extend survival, and modify the immunosuppressive microenvironment. ConclusionsThis investigation elucidated the positive impact of multi-component, multi-target, and multi-approach JPYZXZ intervention for GC treatment.

The Effects of Out-of-School Learning Environments on STEM Education: Teachers’ STEM Awareness and 21st-Century Skills

This study was carried out within the scope of a nature education and science camp project. It aimed to investigate the effects of STEM activities conducted in out-of-school environments on teachers’ STEM awareness and 21st-century skills. The research was designed according to the explanatory sequential pattern of the mixed method, and the quantitative and qualitative methods are the continuation of each other. Twenty-three science, elementary math, technology, and design teachers who worked at public middle schools participated in a six-day program that integrated STEM subjects with out-of-school learning. Seven different STEM-integrated out-of-school activities were implemented for six days in one of the central Anatolian regions in Turkey. STEM awareness and 21st-century skills scales were implemented in the study for data collection. A semi-structured interview was also conducted with teachers. The aim was for the teachers who participated in the project to integrate out-of-school and STEM education and then transfer this newly gained knowledge to their students. Both quantitative and qualitative findings support that teacher’s acquisition of knowledge and skills throughout the science camp would support their students’ craft, analytical thinking, motivation, problem-solving skills, and self-esteem. As such, out-of-learning environments should be used more in both STEM and 21st-century skills education.

The Future of Electronic Commerce in the IoT Environment

The Internet of Things (IoT) was born from the fusion of virtual and physical space and became the initiator of many scientific fields. Economic sustainability is the key to further development and progress. To keep up with the changes, it is necessary to adapt economic models and concepts to meet the requirements of future smart environments. Today, the need for electronic commerce (e-commerce) has become an economic priority during the transition between Industry 4.0 and Industry 5.0. Unlike mass production in Industry 4.0, customized production in Industry 5.0 should gain additional benefits in vertical management and decision-making concepts. The authors’ research is focused on e-commerce in a three-layer vertical IoT environment. The vertical IoT concept is composed of edge, fog, and cloud layers. Given the ubiquity of artificial intelligence in data processing, economic analysis, and predictions, this paper presents a few state-of-the-art machine learning (ML) algorithms facilitating the transition from a flat to a vertical e-commerce concept. The authors also propose hands-on ML algorithms for a few e-commerce types: consumer–consumer and consumer–company–consumer relationships. These algorithms are mainly composed of convolutional neural networks (CNNs), natural language understanding (NLU), sequential pattern mining (SPM), reinforcement learning (RL for agent training), algorithms for clicking on the item prediction, consumer behavior learning, etc. All presented concepts, algorithms, and models are described in detail.

Two adjacent NLR genes conferring quantitative resistance to clubroot disease in Arabidopsis are regulated by a stably inherited epiallelic variation

Clubroot caused by the protist Plasmodiophora brassicae is a major disease affecting cultivated Brassicaceae. Using a combination of quantitative trait locus (QTL) fine mapping, CRISPR-Cas9 validation, and extensive analyses of DNA sequence and methylation patterns, we revealed that the two adjacent neighboring NLR (nucleotide-binding and leucine-rich repeat) genes AT5G47260 and AT5G47280 cooperate in controlling broad-spectrum quantitative partial resistance to the root pathogen P. brassicae in Arabidopsis and that they are epigenetically regulated. The variation in DNA methylation is not associated with any nucleotide variation or any transposable element presence/absence variants and is stably inherited. Variations in DNA methylation at the Pb-At5.2 QTL are widespread across Arabidopsis accessions and correlate negatively with variations in expression of the two genes. Our study demonstrates that natural, stable, and transgenerationally inherited epigenetic variations can play an important role in shaping resistance to plant pathogens by modulating the expression of immune receptors.

MHCpLogics: an interactive machine learning-based tool for unsupervised data visualization and cluster analysis of immunopeptidomes.

The major histocompatibility complex (MHC) encodes a range of immune response genes, including the human leukocyte antigens (HLAs) in humans. These molecules bind peptide antigens and present them on the cell surface for T cell recognition. The repertoires of peptides presented by HLA molecules are termed immunopeptidomes. The highly polymorphic nature of the genres that encode the HLA molecules confers allotype-specific differences in the sequences of bound ligands. Allotype-specific ligand preferences are often defined by peptide-binding motifs. Individuals express up to six classical class I HLA allotypes, which likely present peptides displaying different binding motifs. Such complex datasets make the deconvolution of immunopeptidomic data into allotype-specific contributions and further dissection of binding-specificities challenging. Herein, we developed MHCpLogics as an interactive machine learning-based tool for mining peptide-binding sequence motifs and visualization of immunopeptidome data across complex datasets. We showcase the functionalities of MHCpLogics by analyzing both in-house and published mono- and multi-allelic immunopeptidomics data. The visualization modalities of MHCpLogics allow users to inspect clustered sequences down to individual peptide components and to examine broader sequence patterns within multiple immunopeptidome datasets. MHCpLogics can deconvolute large immunopeptidome datasets enabling the interrogation of clusters for the segregation of allotype-specific peptide sequence motifs, identification of sub-peptidome motifs, and the exportation of clustered peptide sequence lists. The tool facilitates rapid inspection of immunopeptidomes as a resource for the immunology and vaccine communities. MHCpLogics is a standalone application available via an executable installation at: https://github.com/PurcellLab/MHCpLogics.

Context-aware graph embedding with gate and attention for session-based recommendation

Prior solutions on session-based recommendation (SBR) are mainly limited by two major issues: (1) the sequence and transition relationships of items need further integration; (2) the context clues from the neighboring sessions remain largely under-explored. To overcome these issues, this paper proposes a novel Context-aware Graph Embedding Network (CGENet) with gate and attention mechanisms, that not only can effectively exploit the collaborative relationship between the sequence and transition patterns in each session, but also benefit greatly from topological context patterns among different sessions. Specifically, the proposed CGENet consists of three different parts, i.e., Transition Pattern Learning (TPL) module, Sequential Pattern Learning (SPL) module, and Context Pattern Learning (CPL) module. The TPL module is built on Graph Isomorphic Network (GIN) with multiple information highways to capture the transition relationships between items. To maximize the value of sequence-position information, a Gated Multilayer Perceptron (gMLP) is introduced into the SPL module to model the long-term dependencies between sequence tokens. Under the standardized guidance of the graph attention layer, the CPL module can further explore the topological contexts from neighboring sessions, thereby enhancing its ability to predict user preferences more effectively. Extensive experiments on three benchmark datasets demonstrate the superiority of the proposed CGENet compared to the state-of-the-art baselines.

Evidence and emerging trends in local therapy for metastatic hormone-sensitive prostate cancer: a narrative review.

Metastatic hormone-sensitive prostate cancer (mHSPC) displays both simultaneous and sequential patterns of metastasis, emphasizing a comprehensive treatment approach that integrates both local therapy and systemic treatment strategies. The increasing use of molecular imaging has led to a rise in mHSPC diagnoses, underscoring the importance of identifying the right patient population and effective treatment concepts for this disease state. Two prospective trials, HORRAD and STAMP EDE, investigated prostate radiotherapy (RT) for mHSPC; however, they did not show an overall survival (OS) benefit in the unselected cohort. Nonetheless, RT showed favorable outcomes in patients with fewer than five bone metastases, resulting in a 7% 3-year survival improvement and supporting the integration of RT in multimodal treatment for men with oligometastatic mHSPC. Regarding cytoreductive prostatectomy (cRP), the TRoMbone Trial confirmed its feasibility and safety. In addition, findings from the FUSCC-OMPCa Trial demonstrated improved 3-year radiographic progression-free survival and OS rates with acceptable rates of complications and incontinence. Recent data from the LoMP registry have further supported superior OS and cancer-specific survival (CSS) in patients undergoing cRP compared to systemic therapy alone. Notably, no significant differences in OS and CSS were observed between the cRP and RT groups. However, cRP-treated patients exhibited superior 2-year local event-free survival when compared to those treated with RT. RT in combination with systemic therapy remains the established first-line treatment for low-burden mHSPC, though the exact definition of low metastatic burden remains contentious. Precise assessment of metastatic burden is vital to identify patients who would derive the greatest benefit from RT. As treatment paradigms evolve, embracing multimodal approaches holds potential for optimizing outcomes in patients with mHSPC. Further research is needed to solidify the role of cRP as a standard therapeutic approach and to refine treatment strategies for improved patient outcomes.

Identification, comparative and phylogenetic analysis of eight Vitis species based on the chloroplast genome revealed their contribution to heat tolerance in grapevines

The chloroplast genome plays an important role in exploring the origin and evolution of grapes. In this study, the chloroplast genomes of eight grape germplasms were analyzed. The results show that the sequence arrangement was highly conserved and has a typical annular quadripartite structure. At the same time, GC content, gene sequence, repeat type and codon bias pattern are highly conserved. Whereas, 19 mutation hotspot regions were identified by the comparative analysis, including 9 protein-coding genes (atpF, rpoC2, rps18, psbC, atpB, rbcL, rps20, ycf1 and ycf15) and 10 non-coding regions (trnK – rps16, rps16 – trnQ, trnE – trnT, psbZ - trnG, ndhC – trnV, accD – psaI, ycf2 – trnL, ndhF – rpl32, ccsA - ndhD, trnL - ycf2). Most of these highly variable genes were involved in the physiological process of photosynthesis. Further examination of photosynthetic fluorescence showed an obvious difference among those species under high-temperature conditions. In addition, maximum likelihood (ML) evolutionary trees depend on the chloroplast genome and CDS sequences suggest that heat tolerance may be related to the phylogeny of grapevine. In the future, the highly variable sequences and SSR loci were also identified by the chloroplast genome comparison, and those results may be used for the DNA molecular marker in taxonomic studies and contribute to the heat stress resistance breeding of grapevine.

Learning Neighbor User Intention on User–Item Interaction Graphs for Better Sequential Recommendation

The task of sequential recommendation aims to predict a user’s preference by analyzing the user’s historical behaviours. Existing methods model item transitions through leveraging sequential patterns. However, they mainly consider the target user’s behaviours and dynamic characteristics, while often ignoring high-order collaborative connections when modelling user preferences. Some recent works try to use graph-based methods to introduce high-order collaborative signals for sequential recommendation. However, these methods are flawed by two problems: the sequential patterns cannot be effectively mined and their way of introducing high-order collaborative signals is not suitable for sequential recommendation. To address these problems, we propose to fully exploit sequence features and model high-order collaborative signals for sequential recommendation. We propose a N eighbor user I ntention-based S equential Rec ommender (NISRec), which utilizes the intentions of high-order connected neighbor users as high-order collaborative signals in order to improve recommendation performance for the target user. The NISRec contains two main modules: the neighbor user intention embedding module (NIE) and the fusion module. The NIE module describes both the long-term and short-term intentions of neighbor users and aggregates them separately. The fusion module uses these two types of aggregated intentions to model high-order collaborative signals in both the embedding process and user preference modelling phase for recommendations of the target user. Experimental results show that our new approach outperforms the state-of-the-art methods on both sparse and dense datasets. Extensive studies further show the effectiveness of the diverse neighbor intentions introduced by the NISRec.