Distributed representations have been used to support downstream tasks in healthcare recently. Healthcare data (e.g., electronic health records) contain multiple modalities of data from heterogeneous sources that can provide complementary information, alongside an added dimension to learning personalized patient representations. To this end, in this paper we propose a novel unsupervised encoder-decoder model, namely Mixed Pooling Multi-View Attention Autoencoder (MPVAA), that generates patient representations encapsulating a holistic view of their medical profile. Specifically, by first learning personalized graph embeddings pertaining to each patient's heterogeneous healthcare data, it then integrates the non-linear relationships among them into a unified representation through multi-view attention mechanism. Additionally, a mixed pooling strategy is incorporated in the encoding step to learn diverse information specific to each data modality. Experiments conducted for multiple tasks demonstrate the effectiveness of the proposed model over the state-of-the-art representation learning methods in healthcare.
In real-world question-answering (QA) systems, ill-formed questions, such as wrong words, ill word order, and noisy expressions, are common and may prevent the QA systems from understanding and answering them accurately. In order to eliminate the effect of ill-formed questions, we approach the question refinement task and propose a unified model, QREFINE, to refine the ill-formed questions to well-formed question. The basic idea is to learn a Seq2Seq model to generate a new question from the original one. To improve the quality and retrieval performance of the generated questions, we make two major improvements: 1) To better encode the semantics of ill-formed questions, we enrich the representation of questions with character embedding and the recent proposed contextual word embedding such as BERT, besides the traditional context-free word embeddings; 2) To make it capable to generate desired questions, we train the model with deep reinforcement learning techniques that considers an appropriate wording of the generation as an immediate reward and the correlation between generated question and answer as time-delayed long-term rewards. Experimental results on real-world datasets show that the proposed QREFINE method can generate refined questions with more readability but fewer mistakes than the original questions provided by users. Moreover, the refined questions also significantly improve the accuracy of answer retrieval.
Counterfactual thinking describes a psychological phenomenon that people re-infer the possible results with different solutions about things that have already happened. It helps people to gain more experience from mistakes and thus to perform better in similar future tasks. This paper investigates the counterfactual thinking for agents to find optimal decision-making strategies in multi-agent reinforcement learning environments. In particular, we propose a multi-agent deep reinforcement learning model with a structure which mimics the human-psychological counterfactual thinking process to improve the competitive abilities for agents. To this end, our model generates several possible actions (intent actions) with a parallel policy structure and estimates the rewards and regrets for these intent actions based on its current understanding of the environment. Our model incorporates a scenario-based framework to link the estimated regrets with its inner policies. During the iterations, our model updates the parallel policies and the corresponding scenario-based regrets for agents simultaneously. To verify the effectiveness of our proposed model, we conduct extensive experiments on two different environments with real-world applications. Experimental results show that counterfactual thinking can actually benefit the agents to obtain more accumulative rewards from the environments with fair information by comparing to their opponents while keeping high performing efficiency.
Nowadays, with the booming development of the Internet, people benefit from its convenience due to its open and sharing nature. A large volume of natural language texts is being generated by users in various forms, such as search queries, documents, and social media posts. As the unstructured text corpus is usually noisy and messy, it becomes imperative to correctly identify and accurately annotate structured information in order to obtain meaningful insights or better understand unstructured texts. On the other hand, the existing structured information, which embodies our knowledge such as entity or concept relations, often suffers from incompleteness or quality-related issues. Given a gigantic collection of texts which offers rich semantic information, it is also important to harness the massiveness of the unannotated text corpus to expand and refine existing structured knowledge with fewer annotation efforts. In this dissertation, I will introduce principles, models, and algorithms for effective structured knowledge discovery from the massive text corpus. We are generally interested in obtaining insights and better understanding unstructured texts with the help of structured annotations or by structure-aware modeling. Also, given the existing structured knowledge, we are interested in expanding its scale and improving its quality harnessing the massiveness of the text corpus. In particular, four problems are studied in this dissertation: Structured Intent Detection for Natural Language Understanding, Structure-aware Natural Language Modeling, Generative Structured Knowledge Expansion, and Synonym Refinement on Structured Knowledge.
This paper presents a novel framework, MGNER, for Multi-Grained Named Entity Recognition where multiple entities or entity mentions in a sentence could be non-overlapping or totally nested. Different from traditional approaches regarding NER as a sequential labeling task and annotate entities consecutively, MGNER detects and recognizes entities on multiple granularities: it is able to recognize named entities without explicitly assuming non-overlapping or totally nested structures. MGNER consists of a Detector that examines all possible word segments and a Classifier that categorizes entities. In addition, contextual information and a self-attention mechanism are utilized throughout the framework to improve the NER performance. Experimental results show that MGNER outperforms current state-of-the-art baselines up to 4.4% in terms of the F1 score among nested/non-overlapping NER tasks.
Being able to automatically discover synonymous entities from a large free-text corpus has transformative effects on structured knowledge discovery. Existing works either require structured annotations, or fail to incorporate context information effectively, which lower the efficiency of information usage. In this paper, we propose a framework for synonym discovery from free-text corpus without structured annotation. As one of the key components in synonym discovery, we introduce a novel neural network model SynonymNet to determine whether or not two given entities are synonym with each other. Instead of using entities features, SynonymNet makes use of multiple pieces of contexts in which the entity is mentioned, and compares the context-level similarity via a bilateral matching schema to determine synonymity. Experimental results demonstrate that the proposed model achieves state-of-the-art results on both generic and domain-specific synonym datasets: Wiki+Freebase, PubMed+UMLS and MedBook+MKG, with up to 4.16% improvement in terms of Area Under the Curve (AUC) and 3.19% in terms of Mean Average Precision (MAP) compare to the best baseline method.
Being able to recognize words as slots and detect the intent of an utterance has been a keen issue in natural language understanding. The existing works either treat slot filling and intent detection separately in a pipeline manner, or adopt joint models which sequentially label slots while summarizing the utterance-level intent without explicitly preserving the hierarchical relationship among words, slots, and intents. To exploit the semantic hierarchy for effective modeling, we propose a capsule-based neural network model which accomplishes slot filling and intent detection via a dynamic routing-by-agreement schema. A re-routing schema is proposed to further synergize the slot filling performance using the inferred intent representation. Experiments on two real-world datasets show the effectiveness of our model when compared with other alternative model architectures, as well as existing natural language understanding services.
In the era of big data, logistic planning can be made data-driven to take advantage of accumulated knowledge in the past. While in the movie industry, movie planning can also exploit the existing online movie knowledge library to achieve better results. However, it is ineffective to solely rely on conventional heuristics for movie planning, due to a large number of existing movies and various real-world factors that contribute to the success of each movie, such as the movie genre, available budget, production team (involving actor, actress, director, and writer), etc. In this paper, we study a "Blockbuster Planning" (BP) problem to learn from previous movies and plan for low budget yet high return new movies in a totally data-driven fashion. After a thorough investigation of an online movie knowledge library, a novel movie planning framework "Blockbuster Planning with Maximized Movie Configuration Acquaintance" (BigMovie) is introduced in this paper. From the investment perspective, BigMovie maximizes the estimated gross of the planned movies with a given budget. It is able to accurately estimate the movie gross with a 0.26 mean absolute percentage error (and 0.16 for budget). Meanwhile, from the production team's perspective, BigMovie is able to formulate an optimized team with people/movie genres that team members are acquainted with. Historical collaboration records are utilized to estimate acquaintance scores of movie configuration factors via an acquaintance tensor. We formulate the BP problem as a non-linear binary programming problem and prove its NP-hardness. To solve it in polynomial time, BigMovie relaxes the hard binary constraints and addresses the BP problem as a cubic programming problem. Extensive experiments conducted on IMDB movie database demonstrate the capability of BigMovie for an effective data-driven blockbuster planning.
The past few years have witnessed the flourishing of crowdsourced medical question answering (Q&A) websites. Patients who have medical information demands tend to post questions about their health conditions on these crowdsourced Q&A websites and get answers from other users. However, we observe that a large portion of new medical questions cannot be answered in time or receive only few answers from these websites. On the other hand, we notice that solved questions have great potential to solve this challenge. Motivated by these, we propose an end-to-end system that can automatically find similar questions for unsolved medical questions. By learning the vector presentation of unsolved questions and their candidate similar questions, the proposed system outputs similar questions according to the similarity between vector representations. Through the vector representation, the similar questions are found at the question level, and the diversity of medical questions expression issue can be addressed. Further, we handle two more important issues, i.e., training data generation issue and efficiency issue, associated with the LSTM training procedure and the retrieval of candidate similar questions. The effectiveness of the proposed system is validated on a large-scale real-world dataset collected from a crowdsourced maternal-infant Q&A website.