Multimodal Deep Hybrid Representations for Language Generation from Non-textual Data

We aim to develop a framework based on state-of-the-art deep learning models and semantic processing models that learns joint semantic representations from both textual and non-textual data. The system generates crucial semantic information conveyed by [...]

Automatic Intelligent Semantic Processing

We are interested in intelligent processing of data, with a focus on processing of natural language texts. We are particularly interested in the design and development of novel algorithms and models for efficient and [...]

Scalable and Robust Algorithms for Semantic Information Processing for Free Texts

In this proposal, we would like to address the following two practical scientific challenges of semantic processing. First, we focus on the design of highly scalable algorithms for semantic processing. Second, we address the issue [...]

Scalable Models for Deep Semantic Information Processing

In this project, the PI would like to build new semantic processing models based on the success of the PI’s prior research works on efficient semantic parsing. The PI extends such models and algorithms such [...]

Cross-functional Information Systems for Decision Making (CISDEM) – T3: Hidden Knowledge Inference

We design algorithms that exploit security domain knowledge to infer hidden critical knowledge from surface texts, thereby uncovering deeper semantics than possible with existing approaches. We will also build a specialized user interface for presenting [...]

Chinese semantic parsing using compositional vector learning over a dependency based hybrid tree (NSFC No. 61472191)

Semantic parsing is the task of mapping a natural language sentence into a complete, formal meaning representation in a meaning representation language, which is a formal unambiguous language that allows for automated inference and processing. [...]