APPCorp:a corpus for Android privacy policy document structure analysis

With the increasing popularity of mobile devices and the wide adoption of mobile Apps,an increasing concern of privacy issues is raised.Privacy policy is identified as a proper medium to indicate the legal terms,such as the general data protection regulation(GDPR),and to bind legal agreement between service providers and users.However,privacy policies are usually long and vague for end users to read and understand.It is thus important to be able to automatically analyze the document structures of privacy policies to assist user understanding.In this work we create a manually labelled corpus containing 231 privacy policies(of more than 566,000 words and 7,748 annotated paragraphs).We benchmark our data corpus with 3 document classification models and achieve more than 82%on F1-score.

Heuristic solution using decision tree model for enhanced XML schema matching of bridge structural calculation documents

Research on the quality of data in a structural calculation document(SCD)is lacking,although the SCD ofa bridge is used as an essential reference during the entire lifecycle of the facility.XML Schema matching enables qualitative improvement of the stored data.This study aimed to enhance the applicability of XML Schema matching,which improves the speed and quality of information stored in bridge SCDs.First,the authors proposed a method of reducing the computing time for the schema matching of bridge SCDs.The computing speed of schema matching was increased by 13 to 1800 times by reducing the checking process of the correlations.Second,the authors developed a heuristic solution for selecting the optimal weight factors used in the matching process to maintain a high accuracy by introducing a decision tree.The decision tree model was built using the content elements stored in the SCD,design companies,bridge types,and weight factors as input variables,and the matching accuracy as the target variable.The inverse-calculation method was applied to extract the weight factors from the decision tree model for high-accuracy schema matching results.

Document structure model for survey generation using neural network

Survey generation aims to generate a summary from a scientific topic based on related papers.The structure of papers deeply influences the generative process of survey,especially the relationships between sentence and sentence,paragraph and paragraph.In principle,the structure of paper can influence the quality of the summary.Therefore,we employ the structure of paper to leverage contextual information among sentences in paragraphs to generate a survey for documents.In particular,we present a neural document structure model for survey generation.We take paragraphs as units,and model sentences in paragraphs,we then employ a hierarchical model to learn structure among sentences,which can be used to select important and informative sentences to generate survey.We evaluate our model on scientific document data set.The experimental results show that our model is effective,and the generated survey is informative and readable.

Knowledge-driven decision analytics for commercial banking

Although the corporate relationship manager seems to be the key enabler in commercial banking,the personal relationship sales model is not a sustainable model for the paradigm shift in digital financial markets.In this research,we propose a knowledge-driven decision analytics approach to improve the decision process.However,most of the corporate client documents processed in banks are not well-structured and the traditional analysis approach does not consider the document structure,which carries rich semantic information.We propose a document structure-based text representation approach with incorporating auxiliary information in the predictive analytics of unstructured data to improve the performance in the document classification task.The proposed approach significantly outperforms the traditional whole document approach which does not take into considerations of the document structure.With the proposed approach,knowledge can be effectively and efficiently used for business decisions and planning to improve the competitive advantage and substantiality of banks.