Modern data-intensive applications require databases that support fast analytical processing on massive dynamic graphs in real time, while simultaneously providing transactional guarantees for modifying graph-based objects ( i.e. , edges, vertices and their properties). Achieving efficient Hybrid Graph Transactional/Analytical Processing (HGTAP) in a database poses significant challenges due to the simultaneous requirements of high operation throughput, high data freshness, and high performance isolation when processing concurrent read/write queries on intricate graph topology. Existing disk-based graph databases fail to meet these requirements at the same time due to their inclined data layout, such as the transactional storage based on adjacency list and the analytical storage based on CSR (compressed sparse row) format. To address these challenges, we present BACH (Bridging Adjacency List and CSR Format using LSM (Log-Structured Merge)-Trees for HGTAP Workloads) to fill the gaps in HGTAP databases. BACH expands the design space of traditional LSM-Trees to accommodate different graph data layouts in different levels. The compaction process is further extended to seamlessly transform the graph layout from the TP-friendly adjacency list to the AP-friendly CSR format through the data propagation to deeper levels in the LSM-Tree. A novel compaction policy, namely elastic merge , is carefully devised to adapt to diverse workloads and the skew vertex degree distribution on graph data. These techniques lead to a Graph-aware Real-time (GR)-LSM-Tree , which can provide consistently efficient data access for diverse workloads throughout the entire lifespan of graph objects. Then, a lightweight multi-version scheme is devised for the GR-LSM-Tree to accelerate the concurrent read/write processing with the snap-shot isolation guarantee. Comprehensive experiments demonstrate that BACH significantly outperforms other disk-based graph database solutions in HGTAP workloads.

The problem of similarity search in high-dimensional space is a fundamental problem with numerous applications in computer science, yet it remains challenging due to the curse of dimensionality. This paper introduces DForest, a novel indexing approach designed to address this challenge for both range and kNN queries on high-dimensional data. Unlike previous similarity search approaches that apply a fixed dimensionality reduction to all objects uniformly, our approach determines the minimal dimensionality required for each object within a specified loss threshold and then reduces the dimensionality for each object individually. Furthermore, the query performance is also optimized by deriving the upper and lower bounds of retrieved blocks and computing distances in a low-embedding space preferentially. Theoretical analysis is provided to support our search strategy. Extensive experiments on a variety of datasets verify the superiority of DForest over the state-of-the-art methods.

Aspect sentiment triplet extraction is an important task in natural language processing. Previous work tends to focus on the interaction between the aspect and opinion, while ignoring the positive impact of sentiment on interaction within the triplet. In this paper, we propose a novel aspect sentiment triplet extraction model based on dual learning with sequential prompting. This model is designed as a bidirectional extraction framework that fully takes sentiment polarity into account in the interaction process of aspect and opinion. Besides, we introduce a dual loss as a regularization term for the extraction model to promote better learning in both directions. We further design a sequential prompting strategy to determine aspect, opinion, and sentiment polarity more accurately, which utilizes the results extracted in the previous step as prior knowledge to guide the prediction of the next target. We conduct experiments on three public datasets and the results show the effectiveness of our method. More importantly, we deploy our method on Fliggy application and the 14-day online A/B testing indicates that Page View Click-Through Rate and Page View Conversion Rate increase by 1.17% and 1.08% when user short reviews are used for tagging items with the help of our method.

In light of the success of the pre-trained language models (PLMs), continual pre-training of generic PLMs has been the paradigm of domain adaption. In this paper, we propose QUERT, A Continual Pre-trained Language Model for QUERy Understanding in Travel Domain Search. QUERT is jointly trained on four tailored pre-training tasks to the characteristics of query in travel domain search: Geography-aware Mask Prediction, Geohash Code Prediction, User Click Behavior Learning, and Phrase and Token Order Prediction. Performance improvement of downstream tasks and ablation experiment demonstrate the effectiveness of our proposed pre-training tasks. To be specific, the average performance of downstream tasks increases by 2.02% and 30.93% in supervised and unsupervised settings, respectively. To check on the improvement of QUERT to online business, we deploy QUERT and perform A/B testing on Fliggy APP. The feedback results show that QUERT increases the Unique Click-Through Rate and Page Click-Through Rate by 0.89% and 1.03% when applying QUERT as the encoder. Resources are available at https://github.com/hsaest/QUERT

POI tagging aims to annotate a point of interest (POI) with some informative tags, which facilitates many services related to POIs, including search, recommendation, and so on. Most of the existing solutions neglect the significance of POI images and seldom fuse the textual and visual features of POIs, resulting in suboptimal tagging performance. In this paper, we propose a novel M ulti-M odal M odel for P OI T agging, namely M3PT, which achieves enhanced POI tagging through fusing the target POI's textual and visual features, and the precise matching between the multi-modal representations. Specifically, we first devise a domain-adaptive image encoder (DIE) to obtain the image embeddings aligned to their gold tags' semantics. Then, in M3PT's text-image fusion module (TIF), the textual and visual representations are fully fused into the POIs' content embeddings for the subsequent matching. In addition, we adopt a contrastive learning strategy to further bridge the gap between the representations of different modalities. To evaluate the tagging models' performance, we have constructed two high-quality POI tagging datasets from the real-world business scenario of Ali Fliggy. Upon the datasets, we conducted the extensive experiments to demonstrate our model's advantage over the baselines of uni-modality and multi-modality, and verify the effectiveness of important components in M3PT, including DIE, TIF and the contrastive learning strategy.

Chinese abbreviation prediction is very important for various natural language processing tasks such as query understanding and entity linking, since people tend to use the concise abbreviation rather than the full form (name) to mention an entity. The existing models achieve their predictions through sequence labeling, i.e., the binary classification for each character (token) of the full form. However, they only leverage the semantics of the entity itself, overlooking the label dependencies between the tokens, and the rich information of the entity-related texts. In this paper we proposed a Context-Enhanced Transformer with Abbr-Recover policy, namely CETAR, for Chinese abbreviation prediction. CETAR predicts the abbreviation sequence mainly through an iterative decoding process, of which each round consists of an abbreviation and recovery operation. Our extensive experiments upon both general field and specific domain datasets justify that CETAR outperforms the state-of-the-art baselines including sequence labeling models and sequence generation models. Moreover, we have successfully constructed a Chinese abbreviation dataset from the famous tour website Fliggy, and we also shared it at https://github.com/tolerancecky/abbr-0731. The online A/B test on the Fliggy search system shows that 2.03% of conversion rate improvement has been achieved with the predicted abbreviations.

Most massively multiplayer game servers employ static partitioning of their game world into distinct mini-worlds that are hosted on separate servers. This limits cross-server interactions between players, and exposes the division of the world to players. We have designed and implemented an architecture in which the partitioning of game regions across servers is transparent to players and interactions are not limited to objects in a single region or server. This allows a finer grain partitioning, which combined with a dynamic load management algorithm enables us to better handle transient crowding by adaptively dispersing or aggregating regions from servers in response to quality of service violations.Our load balancing algorithm is aware of the spatial locality in the virtual game world. Based on localized information, the algorithm balances the load and reduces the cross server communication, while avoiding frequent reassignment of regions. Our results show that locality aware load balancing reduces the average user response time by up to a factor of 6 compared to a global algorithm that does not consider spatial locality and by up to a factor of 8 compared to static partitioning.