Knowledge graphs have gained popularity for their ability to organize and analyze complex data effectively. When combined with graph embedding techniques, such as graph neural networks (GNNs), knowledge graphs become a potent tool in providing valuable insights. This study explores the application of graph embedding in identifying competitors from a financial knowledge graph. Existing state-of-the-art(SOTA) models face challenges due to the unique attributes of our knowledge graph, including directed and undirected relationships, attributed nodes, and minimal annotated competitor connections. To address these challenges, we propose a novel graph embedding model, JPEC(JPMorgan Proximity Embedding for Competitor Detection), which utilizes graph neural network to learn from both first-order and second-order node proximity together with vital features for competitor retrieval. JPEC had outperformed most existing models in extensive experiments, showcasing its effectiveness in competitor retrieval.

Despite recent progress in deep learning, visual emotion recognition remains a challenging problem due to ambiguity of emotion perception, diverse concepts related to visual emotion and lack of large-scale annotated dataset. In this paper, we present a large-scale multimodal pre-training method to learn visual emotion representation by aligning emotion, object, attribute triplet with a contrastive loss. We conduct our pre-training on a large web dataset with noisy tags and fine-tune on visual emotion classification datasets. Our method achieves state-of-the-art performance for visual emotion classification.

Knowledge Graphs have emerged as a compelling abstraction for capturing key relationship among the entities of interest to enterprises and for integrating data from heterogeneous sources. JPMorgan Chase (JPMC) is leading this trend by leveraging knowledge graphs across the organization for multiple mission critical applications such as risk assessment, fraud detection, investment advice, etc. A core problem in leveraging a knowledge graph is to link mentions (e.g., company names) that are encountered in textual sources to entities in the knowledge graph. Although several techniques exist for entity linking, they are tuned for entities that exist in Wikipedia, and fail to generalize for the entities that are of interest to an enterprise. In this paper, we propose a novel end-to-end neural entity linking model (JEL) that uses minimal context information and a margin loss to generate entity embeddings, and a Wide & Deep Learning model to match character and semantic information respectively. We show that JEL achieves the state-of-the-art performance to link mentions of company names in financial news with entities in our knowledge graph. We report on our efforts to deploy this model in the company-wide system to generate alerts in response to financial news. The methodology used for JEL is directly applicable and usable by other enterprises who need entity linking solutions for data that are unique to their respective situations.

Emotion understanding from an image is an important computer vision research topic, but the reasoning behind the justification has been largely unexplored. We propose to utilize language, and more specifically, explanation of the judgement to increase the explainability for this task. We collect a dataset with image, emotion tag and explanation in natural language, and conduct analysis on the dataset to gain insights on the ambiguity of the human emotion perceptual. We examine baseline methods to predict emotion from image, explanation and general image description, and unifying both modalities. Our experiments shed lights on effects from different modalities and we also identify opportunities for future visual emotion categorization research based on the analysis. We release our dataset to advance future research.

Outfit recommendation requires the answers of some challenging outfit compatibility questions such as 'Which pair of boots and school bag go well with my jeans and sweater?'. It is more complicated than conventional similarity search, and needs to consider not only visual aesthetics but also the intrinsic fine-grained and multi-category nature of fashion items. Some existing approaches solve the problem through sequential models or learning pair-wise distances between items. However, most of them only consider coarse category information in defining fashion compatibility while neglecting the fine-grained category information often desired in practical applications. To better define the fashion compatibility and more flexibly meet different needs, we propose a novel problem of learning compatibility among multiple tuples (each consisting of an item and category pair), and recommending fashion items following the category choices from customers. Our contributions include: 1) Designing a Mixed Category Attention Net (MCAN) which integrates both fine-grained and coarse category information into recommendation and learns the compatibility among fashion tuples. MCAN can explicitly and effectively generate diverse and controllable recommendations based on need. 2) Contributing a new dataset IQON, which follows eastern culture and can be used to test the generalization of recommendation systems. Our extensive experiments on a reference dataset Polyvore and our dataset IQON demonstrate that our method significantly outperforms state-of-the-art recommendation methods.

Flash Sale Sites popularize because they save great money for users. Good recommender systems can further save users' time to improve their online shopping experiences. Although there exist a lot of studies on recommender system, very few focus on flash sale sites. Big Data, Context Sensitivity, and Feature Engineering are three key challenges for one to build a good recommender system. This paper proposes two deep learning oriented models: Tensor-AutoRec and Hybrid-AutoRec to cope with the problems within an industrial context. First, these two models can handle storage and speed problem caused by big data. Second, both models incorporate context information, so they can generate more relevant recommendations by adapting to specific contextual situations. Third, our deep learning-based models can be trained end-to-end without tedious feature engineerings. Extensive experiments with a half year real transcation data demonstrate that our models can outperform classical ones in terms of different evaluation metrices. Finally, online A/B testing results showed that our model can improve our old recommendation system over various online performance indicators.

In this paper, we propose a novel k-anonymization scheme to counter deanonymization queries on social networks. With this scheme, all entities are protected by k-anonymization, which means the attackers cannot re-identify a target with confidence higher than 1/k. The proposed scheme minimizes the modification on original networks, and accordingly maximizes the utility preservation of published data while achieving k-anonymization privacy protection. Extensive experiments on real data sets demonstrate the effectiveness of the proposed scheme, where the efficacy of the k-anonymized networks is verified with the distributions of pagerank, betweenness, and their Kolmogorov-Smirnov (K-S) test.

Understanding ongoing topics and their evolutions in social media is of great importance. Although topic analysis is not a novel research question, social media environment has presented new challenges. First, with insufficient co-occurrence information, short text have undermined many word co-occurrence oriented topic models' applicability. Second, real time message streams make traditional discretized topic tracking methods hard to function. Third, topics' evolution mechanisms are of great importance in social media context, but many studies have ignored them. Forth, topics have more complicated correlation among each other. Considering the existing problems, this paper has proposed a Semi-Supervised Dirichlet-Hawkes Process (SDHP) to deal with topic detection and tracking from social media. The main contributions of this paper are reflected in: (1) SDHP can handle short text problem efficiently; (2) SDHP can track topics from continuous message stream; (3) SDHP can reveal topics' underlying evolution patterns; and (4) SDHP can capture topics' correlations We have evaluated SDHP's ability in both topic detection and tracking in 8 real datasets from Twitter, and the algorithm's performances are very promising.

Video popularity prediction plays a foundational role in many aspects of life, such as recommendation systems and investment consulting. Because of its technological and economic importance, this problem has been extensively studied for years. However, four constraints have limited most related works' usability. First, most feature oriented models are inadequate in the social media environment, because many videos are published with no specific content features, such as a strong cast or a famous script. Second, many studies assume that there is a linear correlation existing between view counts from early and later days, but this is not the case in every scenario. Third, numerous works just take view counts into consideration, but discount associated sentiments. Nevertheless, it is the public opinions that directly drive a video's final success/failure. Also, many related approaches rely on a network topology, but such topologies are unavailable in many situations. Here, we propose a Dual Sentimental Hawkes Process (DSHP) to cope with all the problems above. DSHP's innovations are reflected in three ways: (1) it breaks the "Linear Correlation" assumption, and implements Hawkes Process; (2) it reveals deeper factors that affect a video's popularity; and (3) it is topology free. We evaluate DSHP on four types of videos: Movies, TV Episodes, Music Videos, and Online News, and compare its performance against 6 widely used models, including Translation Model, Multiple Linear Regression, KNN Regression, ARMA, Reinforced Poisson Process, and Univariate Hawkes Process. Our model outperforms all of the others, which indicates a promising application prospect.

Cocitation and co‐word methods have long been used to detect and track emerging topics in scientific literature, but both have weaknesses. Recently, while many researchers have adopted generative probabilistic models for topic detection and tracking, few have compared generative probabilistic models with traditional cocitation and co‐word methods in terms of their overall performance. In this article, we compare the performance of hierarchical Dirichlet process (HDP), a promising generative probabilistic model, with that of the 2 traditional topic detecting and tracking methods—cocitation analysis and co‐word analysis. We visualize and explore the relationships between topics identified by the 3 methods in hierarchical edge bundling graphs and time flow graphs. Our result shows that HDP is more sensitive and reliable than the other 2 methods in both detecting and tracking emerging topics. Furthermore, we demonstrate the important topics and topic evolution trends in the literature of terrorism research with the HDP method.

Sentiment analysis studies the public opinions towards an entity, and it is an important research area in data mining. Recently, a lot of sentiment analysis models have been proposed, including supervised and unsupervised approaches. However, the role of supervised models has been undermined by the phenomenon of big data, and the unsupervised ones are drawing more and more attention. But, most current unsupervised methods are based on Latent Dirichlet Allocation (LDA), and they need to specify the number of aspects in advance, making them subjective. In addition, these methods treat factual words and opinioned words the same, and assume that one sentence contains only one aspect, all of which make the existing unsupervised methods unsatisfactory. To solve these problems, this paper proposes a novel hybrid Hierarchical Dirichlet Process-Latent Dirichlet Allocation (HDP-LDA) model. This model can automatically determine the number of aspects, distinguish factual words from opinioned words, and further effectively extracts the aspect specific sentiment words. Experiment result shows that our model can clearly capture the aspects people mentioned and the specific sentiment words they use in each aspect, improving the performance of sentiment analysis efficiently. At last, we compared our model with the influential topic models, namely, JST, AUSM and Maxine-LDA, on the online restaurant review, and found our model performs very well.