Item ID Research Articles

Personalized Explainable Recommendations for Self-Attention Collaboration

In recommender systems, providing reasonable explanations can enhance users’ comprehension of recommended results. Template-based explainable recommendation heavily relies on pre-defined templates, constraining the expressiveness of generated sentences and resulting in low-quality explanations. Recently, a novel approach was introduced, utilizing embedding representations of items and comments to address the issue of user IDs and item IDs not residing in the same semantic space as words, thus attributing linguistic meaning to IDs. However, these models often fail to fully exploit collaborative information within the data. In personalized recommendation and explanation processes, understanding the user’s emotional feedback and feature preferences is paramount. To address this, we propose a personalized explainable recommendation model based on self-attention collaboration. Initially, the model employs an attention network to amalgamate the user’s historical interaction feature preferences with their user ID information, while simultaneously integrating all feature information of the item with its item ID to enhance semantic ID representation. Subsequently, the model incorporates the user’s comment feature rhetoric and sentiment feedback to generate more personalized recommendation explanations utilizing a self-attention network. Experimental evaluations conducted on two datasets of varying scales demonstrate the superiority of our model over current state-of-the-art approaches, validating its effectiveness.

Asymmetric multilevel interactive attention network integrating reviews for item recommendation

Recently, most studies in the field have focused on integrating reviews behind ratings to improve recommendation performance. However, two main problems remain (1) Most works use a unified data form and the same processing method to address the user and the item reviews, regardless of their essential differences. (2) Most works only adopt simple concatenation operation when constructing user-item interaction, thus ignoring the multilevel relationship between the user and the item, which may lead to suboptimal recommendation performance. In this paper, we propose a novel Asymmetric Multi-Level Interactive Attention Network (AMLIAN) integrating reviews for item recommendation. AMLIAN can predict precise ratings to help the user make better and faster decisions. Specifically, to address the essential difference between the user and the item reviews, AMLIAN uses the asymmetric network to construct user and item features using different data forms (document-level and review-level). To learn more personalized user-item interaction, the user ID and item ID and some processed features of user reviews and item reviews are respectively used for multilevel relationships. Experiments on five real-world datasets show that AMLIAN significantly outperforms state-of-the-art methods.

Dual-View Whitening on Pre-trained Text Embeddings for Sequential Recommendation

Recent advances in sequential recommendation models have demonstrated the efficacy of integrating pre-trained text embeddings with item ID embeddings to achieve superior performance. However, our study takes a unique perspective by exclusively focusing on the untapped potential of text embeddings, obviating the need for ID embeddings. We begin by implementing a pre-processing strategy known as whitening, which effectively transforms the anisotropic semantic space of pre-trained text embeddings into an isotropic Gaussian distribution. Comprehensive experiments reveal that applying whitening to pre-trained text embeddings in sequential recommendation models significantly enhances performance. Yet, a full whitening operation might break the potential manifold of items with similar text semantics. To retain the original semantics while benefiting from the isotropy of the whitened text features, we propose a Dual-view Whitening method for Sequential Recommendation (DWSRec), which leverages both fully whitened and relaxed whitened item representations as dual views for effective recommendations. We further examine the advantages of our approach through both empirical and theoretical analyses. Experiments on three public benchmark datasets show that DWSRec outperforms state-of-the-art methods for sequential recommendation.

Multimodal representation learning for tourism recommendation with two-tower architecture.

Personalized recommendation plays an important role in many online service fields. In the field of tourism recommendation, tourist attractions contain rich context and content information. These implicit features include not only text, but also images and videos. In order to make better use of these features, researchers usually introduce richer feature information or more efficient feature representation methods, but the unrestricted introduction of a large amount of feature information will undoubtedly reduce the performance of the recommendation system. We propose a novel heterogeneous multimodal representation learning method for tourism recommendation. The proposed model is based on two-tower architecture, in which the item tower handles multimodal latent features: Bidirectional Long Short-Term Memory (Bi-LSTM) is used to extract the text features of items, and an External Attention Transformer (EANet) is used to extract image features of items, and connect these feature vectors with item IDs to enrich the feature representation of items. In order to increase the expressiveness of the model, we introduce a deep fully connected stack layer to fuse multimodal feature vectors and capture the hidden relationship between them. The model is tested on the three different datasets, our model is better than the baseline models in NDCG and precision.

Hybrid Prompt Recommendation Explanation Generation combined with Graph Encoder

Recommendation systems have been effectively utilized in various fields, but their internal decision-making methods are still largely unknown. This opaque decision-making method can greatly affect users’ trust in the recommendation system. Therefore, finding a way to explain the reasons for model decisions has become an urgent task. Previous studies often used LSTM and other models to generate recommendation explanations and explain the reasons for recommendations in text form. However, traditional methods cannot effectively use the ID information of users and items, and the text generated is highly repetitive. To solve this problem, this paper uses the method of prompt learning combined with a graph encoder to design a recommendation explanation generation model. In order to narrow the semantic gap between the ID information of users and items and natural language and capture high-level interaction information, this paper designs a graph encoder based on user similarity to learn the interactive semantic information of user and item IDs, and to construct a continuous prompt. Then, the discrete prompt composed of discrete features of users and items is combined with the continuous prompt to construct a hybrid prompt to input into the pre-trained model to generate the recommended explanation. This paper experiments on three publicly available datasets and compares them with several state-of-the-art methods to demonstrate the personalization and text quality of the generated explanations.

GBIF-Compliant Data Pipeline for the Management and Publication of a Global Taxonomic Reference List of Pests in Natural History Collections

There is a growing demand for monitoring pests in natural history collections (NHCs) and establishing integrated pest management (IPM) solutions (Crossman and Ryde 2022). In this context, up-to-date taxonomic reference lists and controlled vocabularies following standard schemes are crucial and facilitate recording organisms detected in collections. The data pipeline described here results in the publication of a taxon reference list based on information from online resources and standard IPM literature. Most of the over 140 pest taxa on species level and above are insects, the rest belong to other animal groups and fungi. The complete taxon names, synonyms, English and German common names, and the hierarchical classification (parent-child relationships) are organised in a client-server installation of DiversityTaxonNames (DTN) at the Bavarian Natural History Collections (SNSB). DTN is a Microsoft Structured Query Language (MS SQL) database tool of the Diversity Workbench (DWB) framework with a published Entity Relation (ER) diagram (Hagedorn et al. 2019). The management is done using the Global Biodiversity Information Facility (GBIF) backbone taxonomy as external name resource, with linkage to the respective Wikidata Q item ID as a external persistent identifier (PID). Moreover, information on pest occurrence in NHCs is given, distinguishing the Consortium of European Taxonomic Facilities (CETAF) major NHC collection types affected (i.e., heritage sciences, life sciences and earth sciences) and the object categories, e.g., natural objects/specimens damaged. The data management in DTN enables the long-running curation, done by list curators. The generic data pipeline for the management and publication of a Global Taxonomic Reference List of Pests in NHCs is based on the DTN taxon lists concept and architecture and described under About "Taxon list of pest organisms for IPM at natural history collections compiled at the SNSB". It includes four steps (A–D) with significant results for best practices of data processing (Fig. 1). A. The data is managed and processed for publication by list curators in the database DiversityTaxonNames (DTN). As a result, the list can be kept up-to-date and is—without transformation—ready to be used for IPM solutions at any NHC with a DiversityCollection installation and as part of the DWB cloud services. B. The up-to-date data is publicly available via the DTN REST Webservice for Taxon Lists with machine-readable Application Programming Interface (API). As a result, the dynamic list publication service can be used as a reference backbone for establishing IPM solutions for pest monitoring at any NHC. C. The data is provided via the GBIF checklist data publication pipeline of the SNSB through GBIF validation tools and Darwin Core Archive in DwC-A (zip format) for GBIF. As a result, the checklist information becomes part of the GBIF network with GBIF ChecklistBank and GBIF Global Taxonomy. This ensures future compliance of data with the Findability, Accessibility, Interoperability, and Reuse (FAIR) guiding principles. D. The DTN REST Web service for Taxon Lists (currently 60 lists) is registered and accessible through the German Federation for Biological Data (GFBio) Terminology service. As a result, the lists with external PIDs and other information are available as a service (see DTN lists overview). In the upcoming Research Data Commons of the German National Research Data Infrastructure (NFDI) Initiative (Diepenbroek et al. 2021), it will be part of a standardized layer of APIs with an agreed interface scheme for improved accessibility. The provided tools, API and data are part of the upcoming NFDI4Biodiversity service portfolio. Future scenarios include the use of the list items and properties as classes for diagnosis purposes with DiversityNaviKey (Triebel et al. 2021) including the publication of images for identifying pests.

Personalized Prompt Learning for Explainable Recommendation

Providing user-understandable explanations to justify recommendations could help users better understand the recommended items, increase the system’s ease of use, and gain users’ trust. A typical approach to realize it is natural language generation. However, previous works mostly adopt recurrent neural networks to meet the ends, leaving the potentially more effective pre-trained Transformer models under-explored. In fact, user and item IDs, as important identifiers in recommender systems, are inherently in different semantic space as words that pre-trained models were already trained on. Thus, how to effectively fuse IDs into such models becomes a critical issue. Inspired by recent advancement in prompt learning, we come up with two solutions: find alternative words to represent IDs (called discrete prompt learning) and directly input ID vectors to a pre-trained model (termed continuous prompt learning). In the latter case, ID vectors are randomly initialized but the model is trained in advance on large corpora, so they are actually in different learning stages. To bridge the gap, we further propose two training strategies: sequential tuning and recommendation as regularization. Extensive experiments show that our continuous prompt learning approach equipped with the training strategies consistently outperforms strong baselines on three datasets of explainable recommendation.

Learning the User’s Deeper Preferences for Multi-modal Recommendation Systems

Recommendation system plays an important role in the rapid development of micro-video sharing platform. Micro-video has rich modal features, such as visual, audio, and text. It is of great significance to carry out personalized recommendation by integrating multi-modal features. However, most of the current multi-modal recommendation systems can only enrich the feature representation on the item side, while it leads to poor learning of user preferences. To solve this problem, we propose a novel module named Learning the User’s Deeper Preferences (LUDP) , which constructs the item-item modal similarity graph and user preference graph in each modality to explore the learning of item and user representation. Specifically, we construct item-item similar modalities graph using multi-modal features, the item ID embedding is propagated and aggregated on the graph to learn the latent structural information of items; The user preference graph is constructed through the historical interaction between the user and item, on which the multi-modal features are aggregated as the user’s preference for the modal. Finally, combining the two parts as auxiliary information enhances the user and item representation learned from the collaborative signals to learn deeper user preferences. Through a large number of experiments on two public datasets (TikTok, Movielens), our model is proved to be superior to the most advanced multi-modal recommendation methods.

Click-through rate prediction model integrating user interest and multi-head attention mechanism

The purpose of click-through rate (CTR) prediction is to anticipate how likely a person is to click on an advertisement or item. It's required for a lot of internet applications, such online advertising and recommendation systems. The previous click-through rate estimation approach suffered from the following two flaws. On the one hand, input characteristics (such as user id, user age, user age, item id, item category) are usually sparse and multidimensional, making them effective. High-level combination characteristics are used for prediction. Obtaining it manually by domain experts takes a long time and is difficult to finish; also, customer interests are not all the same. The accuracy of the model findings will significantly increase if this immediately recognized component is incorporated in the prediction model. As a consequence, this study creates an IARM (interactive attention rate estimation model) that incorporates user interest as well as a multi-head self-attention mechanism. The deep learning network is used in the model to determine the user's interest expression based on user attributes. The multi-head self-attention mechanism with residual network is then employed to get feature interaction, which enhances the degree of effect of significant characteristics on the estimation result as well as its accuracy. The IARM model outperforms other recent prediction models in the assessment metrics AUC and LOSS, and it has superior accuracy, according to the results from the public experimental data set.

A time-aware self-attention based neural network model for sequential recommendation

Sequential recommendation is one of the hot research topics in recent years. Various sequential recommendation models have been proposed, of which Self-Attention (SA)-based models are shown to have state-of-the-art performance. However, most of the existing SA-based sequential recommendation models do not make use of temporal information, i.e., timestamps of user–item interactions, except for an initial attempt (Li et al., 2020). In this paper, we propose a Time-Aware Transformer for Sequential Recommendation (TAT4SRec), an SA-based neural network model which utilizes the temporal information and captures users’ preferences more precisely. TAT4SRec has two salient features: (1) TAT4SRec utilizes an encoder–decoder structure to model timestamps and interacted items separately and this structure appears to be a better way of making use of the temporal information. (2) in the proposed TAT4SRec, two different embedding modules are designed to transform continuous data (timestamps) and discrete data (item IDs) into embedding matrices respectively. Specifically, we propose a window function-based embedding module to preserve the continuous dependency contained in similar timestamps. Finally, extensive experiments demonstrate the effectiveness of the proposed TAT4SRec over various state-of-the-art MC/RNN/SA-based sequential recommendation models under several widely-used metrics. Furthermore, experiments are also performed to show the rationality of the different proposed structures and demonstrate the computation efficiency of TAT4SRec. The promising experimental results make it possible to apply TAT4SRec in various online applications.

Neural Collaborative with Sentence BERT for News Recommender System

The number of news produced every day is as much as 3 million per day, making readers have many choices in choosing news according to each reader's topic and category preferences. The recommendation system can make it easier for users to choose the news to read. The method that can be used in providing recommendations from the same user is collaborative filtering. Neural collaborative filtering is usually being used for recommendation systems by combining collaborative filtering with neural networks. However, this method has the disadvantage of recommending the similarity of news content such as news titles and content to users. This research wants to develop neural collaborative filtering using sentences BERT. Sentence BERT is applied to news titles and news contents that are converted into sentence embedding. The results of this sentence embedding are used in neural collaboration with item id, user id, and news category. We use a Microsoft news dataset of 50,000 users and 51,282 news, with 5,475,542 interactions between users and news. The evaluation carried out in this study uses precision, recall, and ROC curves to predict news clicks by the user. Another evaluation uses a hit ratio with the leave one out method. The evaluation results obtained a precision value of 99.14%, recall of 92.48%, f1-score of 95.69%, and ROC score of 98%. Evaluation measurement using the hit ratio@10 produces a hit ratio of 74% at fiftieth epochs for neural collaborative with sentence BERT which is better than neural collaborative filtering (NCF) and NCF with news category.

Auxiliary Information-Enhanced Recommendations

Sequential recommendations have attracted increasing attention from both academia and industry in recent years. They predict a given user’s next choice of items by mainly modeling the sequential relations over a sequence of the user’s interactions with the items. However, most of the existing sequential recommendation algorithms mainly focus on the sequential dependencies between item IDs within sequences, while ignoring the rich and complex relations embedded in the auxiliary information, such as items’ image information and textual information. Such complex relations can help us better understand users’ preferences towards items, and thus benefit from the recommendations. To bridge this gap, we propose an auxiliary information-enhanced sequential recommendation algorithm called memory fusion network for recommendation (MFN4Rec) to incorporate both items’ image and textual information for sequential recommendations. Accordingly, item IDs, item image information and item textual information are regarded as three modalities. By comprehensively modelling the sequential relations within modalities and interaction relations across modalities, MFN4Rec can learn a more informative representation of users’ preferences for more accurate recommendations. Extensive experiments on two real-world datasets demonstrate the superiority of MFN4Rec over state-of-the-art sequential recommendation algorithms.

Visual content-enhanced sequential recommendation with feature-level attention

A sequential recommender system (SRS) takes historical user-item interactions as a sequence to predict the next item that may be of interest to a user. Most of existing SRSs make predictions by only considering item IDs, while ignoring other important factors that may also significantly influence users’ choices, such as the visual content of items. This may not always be true in reality since an item’s appearance usually has great visual impact on users and thus plays an important role in users’ choices. In practice, each item’s appearance has often been specifically designed to represent a particular style or look. In addition, users’ preferences with regard to visual aspects, such as styling, color, etc., may be relatively consistent compared with their dynamic preferences on items for a certain period under sequential transaction scenarios. This motivates us to take the items’ visual content into account to build more reliable SRSs. Accordingly, we propose a novel visual content-enhanced sequential recommender system (VCSRS) for improving the performance of sequential recommendations. Particularly, in VCSRS, a feature-level attention module (FAM) is designed to learn the attentive visual representations of an item’s appearance. Moreover, a vision-concentrated recurrent network (VCRN) is devised to model the sequential dependencies between items while incorporating items’ visual representations. Extensive experiments on real-world datasets demonstrated the effectiveness of visual information and showed the superiority of our approach over other representative and state-of-the-art methods.

An interpretable mechanism for personalized recommendation based on cross feature

The existing recommender system provides personalized recommendation service for users in online shopping, entertainment, and other activities. In order to improve the probability of users accepting the system’s recommendation service, compared with the traditional recommender system, the interpretable recommender system will give the recommendation reasons and results at the same time. In this paper, an interpretable recommendation model based on XGBoost tree is proposed to obtain comprehensible and effective cross features from side information. The results are input into the embedded model based on attention mechanism to capture the invisible interaction among user IDs, item IDs and cross features. The captured interactions are used to predict the match score between the user and the recommended item. Cross-feature attention score is used to generate different recommendation reasons for different user-items.Experimental results show that the proposed algorithm can guarantee the quality of recommendation. The transparency and readability of the recommendation process has been improved by providing reference reasons. This method can help users better understand the recommendation behavior of the system and has certain enlightenment to help the recommender system become more personalized and intelligent.

Deep Interest-Shifting Network with Meta-Embeddings for Fresh Item Recommendation

Nowadays, people have an increasing interest in fresh products such as new shoes and cosmetics. To this end, an E-commerce platform Taobao launched a fresh-item hub page on the recommender system, with which customers can freely and exclusively explore and purchase fresh items, namely, the New Tendency page. In this work, we make a first attempt to tackle the fresh-item recommendation task with two major challenges. First, a fresh-item recommendation scenario usually faces the challenge that the training data are highly deficient due to low page views. In this paper, we propose a deep interest-shifting network (DisNet), which transfers knowledge from a huge number of auxiliary data and then shifts user interests with contextual information. Furthermore, three interpretable interest-shifting operators are introduced. Second, since the items are fresh, many of them have never been exposed to users, leading to a severe cold-start problem. Though this problem can be alleviated by knowledge transfer, we further babysit these fully cold-start items by a relational meta-Id-embedding generator (RM-IdEG). Specifically, it trains the item id embeddings in a learning-to-learn manner and integrates relational information for better embedding performance. We conducted comprehensive experiments on both synthetic datasets as well as a real-world dataset. Both DisNet and RM-IdEG significantly outperform state-of-the-art approaches, respectively. Empirical results clearly verify the effectiveness of the proposed techniques, which are arguably promising and scalable in real-world applications.

Attentive capsule network for click-through rate and conversion rate prediction in online advertising

Estimating Click-through Rate (CTR) and Conversion Rate (CVR) are two essential user response prediction tasks in computing advertising and recommendation systems. The mainstream methods map sparse, high-dimensional categorical features (e.g., user id, item id) into low-dimensional representations with neural networks. Although they have achieved significant advancement in recent years, how to capture user’s diverse interests effectively from past behaviors is still challenging. Recently some works try using attention-based methods to learn the representation from user behavior history adaptively. However, it is insufficient to capture the diversity of user’s interests. As a step forward to improve this goal, we propose a method named Attentive Capsule Network (ACN). It uses Transformers for feature interaction and leverages capsule networks to capture multiple interests from user behavior history. To precisely obtain sequence representation related to the current advertisement, we further design a modified dynamic routing algorithm integrating with an attention mechanism. Experimental results on real-world datasets demonstrate the effectiveness of our proposed ACN with significant improvement over state-of-the-art approaches. Moreover, it also offers good explainability when extracting diverse interest points of users from behavior history.

Finding Persistent Items in Distributed Datasets

This paper concerns the problem of finding persistent items in distributed datasets, which has many applications such as port scanning and intrusion detection. To the best of our knowledge, there is no existing solution for finding persistent items in distributed datasets. In this paper, we propose DISPERSE, a probabilistic algorithm that can find persistent items in distributed datasets without collecting all the datasets. Our basic idea is that each monitor compresses each item ID in its dataset in a lossy fashion, sends the set of lossily compressed item IDs to the server, then the server recovers the IDs of the persistent items. We design the lossy compression so that given one lossily compressed item ID, the server cannot recover the item ID, but when the number of lossily compressed versions of the same item ID exceeds a threshold, which means that such items are persistent ones, the server can recover the item ID with a high probability. This threshold is exactly the threshold in the definition of persistent items. We implemented DISPERSE and evaluated its performance. In comparison with the straightforward solution, DISPERSE achieves a compression ratio of 26.5% with FNR=3.5% and FPR=0. In comparison with a developed Bloom filter based scheme and the adapted kBF and IBF schemes, our scheme can achieve 7.9, 5.7, and 6.6 times performance gains, respectively, in terms of compression ratio.

Vehicle Recognition based on Support Vector Machine

These days, there is a colossal progression in zones of computerization and PC vision. Item ID is a basic procedure in these innovations. It distinguishes a particular item from a picture or video arrangement and the move is made in like manner. AI calculations are widely utilized for article ID in different applications. The essential highlights are removed from the pictures and are prepared utilizing different classifiers. This paper proposes an article recognizable proof method utilizing Support Vector Machines (SVM). The proposed framework is contrasted and Decision Tree (DT) and K-Nearest Neighbor (KNN) characterization calculations. The item ID framework is surveyed on ID precision, prevision and review.

A deep neural networks based recommendation algorithm using user and item basic data

User basic data (e.g. user gender, user age and user ID, etc.) and item basic data (e.g. item name, item category, etc.) are important side data that can be used to enhance the performance of recommendation algorithms, whereas attempts concerning this issue are still relatively scarce. In this study, a deep neural networks based recommendation algorithm is proposed where user average rating, user basic data (user gender, user age, user occupation, user ID), item basic data (item name, item category, item ID) and item average rating are used. The main idea of the algorithm is to build a regression model for predicting user ratings based on deep neural networks. For this, according to the user data (user average rating and user basic data) and the item data (items basic data and item average rating), a user feature matrix and an item feature matrix are respectively constructed using the four types of neural network layers [i.e., embedding layer (EL), convolution layer (CL), pooling layer (PL) and fully connected layer (FCL)]. Then, based on the obtained user feature matrix and item feature matrix, a user-item feature matrix is further constructed using a FCL. On this basis, a regression model for predicting user ratings can be trained, and a recommendation list can be generated according to the predicted user ratings. To verify the effectiveness of the proposed algorithm, three experiments are conducted using the real data from the MovieLens website. The results of experiments show that the proposed algorithm not only outperforms the state-of-the-art collaborative filtering (CF) recommendation algorithms but also alleviates the data sparsity problem and cold-start problem that would occur when the state-of-the-art CF recommendation algorithms are used.

On the Cover

It is a common practice to track test forms in equating chains across various administrations. The submitted plot aims to augment form information in an equating chain diagram to help inform and evaluate assessment and equating designs. In general, the plot displays a chronological sequence of forms that have been administered, and form information including 1) form name, 2) form difficulty, 3) administration time, 4) cumulative test-taker volume on the form, 5) problematic item information (i.e., DNS as “Do not score” item), 6) DNS item id, and 7) whether the form has been reused. The plot resembles a tree where each circle's location defines the average raw score for a particular form at a given administration. The line connections indicate the form sequence in the equating chains. The plot shows the spread of the average raw scores across the forms within each administration in relation to the scaling form, and over administrations which can help detect any consistent shift over time (e.g., forms getting easier). The plot can also inform the potential overuse of any form that has been reused and has a cumulative test-taker volume up to its most recent administration. In addition, the plot offers dynamic features to 1) show form information once the mouse is hovered over any circle, and 2) enable users to zoom in and pan the plot so that the form information can be scrutinized easily. If you are interested in learning more about this informative and detailed data visualization, contact the principal author Caiyan Zhang (cazhang@collegeboard.org). We want to hear your feedback! Let us know what you think by emailing Ally Shay Thomas (thomasa11@upmc.edu).