Xueqing ("Susan") Liu

Office: 2113 Siebel Center
Computer Science Department
University of Illinois Urbana-Champaign
201 N Goodwin Ave
Urbana IL

Email: xliu93 AT illinois DOT edu

I am a Ph.D. candidate at University of Illinois Urbana-Champaign (Computer Science Department). I am fortunate to be advised by Prof. Chengxiang (“Cheng”) Zhai and co-advised by Prof. Tao Xie.

Prior to UIUC, I received my B.S. degree from the Institution of Interdesciplinary Information Sciences (Tsinghua Xuetang Program Directed by Prof. Andrew Yao) in Tsinghua University in China, 2012.

Research Interests

  • Data science, human factors, text mining, applied machine learning, software engineering, privacy
  • My research interests are user-centered and data-driven. They include two parts: First, designing machine learning approaches to assist users with difficult interactive tasks (i.e., learning to interact with users to solve their ``real pains''). Specifically, I have been working on e-Commerce search (assisting online shoppers to quickly find an item), software development question retrieval (assisting developers find semantically relevant questions), and mobile security decision making (assisting users understand the purpose of data collection). My recent ongoing work focuses on natural language to code synthesis, i.e., semantic parsing. Second, I am also interested in designing statistical studies for discovering new insights in human factors. More details can be found from my research statement.
  • I am on the job market this year, please check out my [Research Statement] [Teaching Statement]. Below are my interview schedules.
  • Projects

    Text Mining for Mobile Security Interaction

      To protect user privacy, it is important for business owners to know that users are aware of the purposes of data processing and their decisions on data processing are consensual. Such importance is strengthened by the recent European General Data Protection Regulation (GDPR). In this project, we conduct a large scale measurement study on how clearly Android apps (under Android 6.0 or higher) explain the purposes of data processing during runtime permission requests. We find that despite many apps provided explanations, the proportion of apps explaining at least one permission is relatively low, the apps also focus on explaining straightforward permissions and neglect difficult ones, further more, a significant proportion of such explanations incorrectly described the actually requested permissions in the Android manifest file.

    CLAP: A Recommender System for Assisting User Security Interaction

      CLAP (CoLlaborative App Permission Explanations) is a recommender system that helps Android app developers improve the clarity of their explanations for the permissions requested by their apps. Given the basic information of the app (title, description, requested permissions and the permission to explain), CLAP automatically suggest a list of sentences by retrieving such sentences from similar apps using the same permission. In addition to finding similar apps, the retrieval algorithm leverages permission-description detection, syntactic structure, text summarization and truth finding techniques to enhance the quality of recommended sentence list. Developers can leverage the ranked list to incrementally revise their existing explanations, or adopt a sentence directly.

    Learning Search Log to Optimize Numerical Facet Interface

      Online marketplaces such as Taobao and Amazon sells products by third-party merchandisers. This results in the challenge to organize the different products specification by different merchandisers. Further more, with the frequently diversified ranking results, the top ranked results are often so messy that processing such information requires significant cognitive overload from users. For example, among the top-20 results of "iPhone Xs" on Taobao, there are 19 different prices (with standard deviation=200$). In this project, we propose to systematically study the new research problem of numerical facet range partition to help organize information for browsing purposes. Our system is adaptive to range number, adaptive to query and user shopping behaviors. We find the optimization problem can be reduced to minimizing its approximate upper bound which largely reduces the time complexity.

    Interactive Hierarchical Moment-based Inference

      Constructing a topic hierarchy for large text collection, such as business documents, news articles, social media messages, and research publications, is helpful for information workers, data analysts and researchers to summarize and navigate them in multiple granularity efficiently. However, complete automatic approaches are often error prone, often failing to meet user requirements. We proposes to give users freedom to construct topical hierarchies via interactive operations such as expanding a branch and merging several branches. We build our approach based on a spectral learning framework named moment-based inference method, and our technical contributions are of two folds. First, we derive robust inference solutions for each operation, so that user editing does not lose information for the inference. Second, we optimize the algorithms of moment-based framework, so our proposed method is orders of magnitude faster than existing hierarchical topic construction methods.


    • [Web Search and Information Retrieval]

      • LinkSO: A Benchmark for Learning to Retrieve Similar Question-Answer Pairs on Software Development Forums
        X. Liu, Chi Wang, Yue Leng, ChengXiang Zhai.
        In ESEC/FSE Workshop on NLP for Software Engineering (NL4SE 2018),
        Lake Buena Vista, Florida, November 2018
        [Short Paper] [Dataset]

      • Information Retrieval Evaluation as Search Simulation: A General Formal Framework for IR Evaluation
        Yinan Zhang, X. Liu and ChengXiang Zhai.
        In Proceedings of the International Conference on the Theory of Information Retrieval (ICTIR 2017),
        Amsterdam, Netherland, October 2017.

        [Paper], [Demo]

      • Numerical Facet Range Partition: Evaluation Metric and Methods
        X. Liu, ChengXiang Zhai, Wei Han and Onur Gungor
        In Proceedings of the International on World Wide Web Conference (19.6%) (WWW 2017),
        Industry Track, Perth, Australia, April 2017.

        [Paper], [Slides]

      • User Fatigue in Online News Recommendation
        Hao Ma, X. Liu, Zhihong Shen
        In Proceedings of the International on World Wide Web Conference (17.3%) (WWW 2016),
        Industry Track, Montreal, Canada, April 2016.


    • [Text Mining for Mobile Security Interactions]

      • A Large-scale Empirical Study on Android Runtime-Permission Rationales
        X. Liu, Yue Leng, Wei Yang, Wenyu Wang, ChengXiang Zhai and Tao Xie.
        In Proceedings of the IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC 2018), Lisbon, Portugal, October 2018.
        [Paper], [Dataset], [Slides]

      • Mining Android App Descriptions for Permission Requirements Recommendation
        X. Liu, Yue Leng, Wei Yang, ChengXiang Zhai and Tao Xie.
        In Proceedings of the IEEE International Requirements Engineering Conference (22.2%) (RE 2018),
        Banff, Canada, August 2018.

        [Paper], [Dataset], [Slides]

    • [Interactive Data Exploration of Text Corpus]

      • Towards Interactive Construction of Topical Hierarchy: A Recursive Tensor Decomposition Approach
        Chi Wang, X. Liu, Yanglei Song, Jiawei Han
        In Proceedings of the SIGKDD Conference on Knowledge Discovery and Data Mining (19.5%) (KDD 2015),
        Sydney, Australia, August 2015.

        [Paper], [Code]

      • Scalable Moment-based Inference for Latent Dirichlet Allocation
        Chi Wang, X. Liu, Yanglei Song, Jiawei Han
        In Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases (23.8%) (ECML/PKDD 2014)
        Nancy, France, September 2014.


      • Scalable Exact Inference for Topic Model
        Chi Wang, X. Liu, Yanglei Song, Jiawei Han
        In ICML Workshop on Moment Based Inference,
        Beijing, China, 2014.

    • [Taxonomy Construction of Text Corpus]

      • Automatic Taxonomy Construction from Keywords via Scalable Bayesian Rose Trees
        Yangqiu Song, Shixia Liu, X. Liu, Haixun Wang
        In IEEE Transactions on Knowledge and Data Engineering (TKDE)
        Volume 27(7), Pages 1861 - 1874, July 2015.


      • Automatic Taxonomy Construction from Keywords
        X. Liu, Yangqiu Song, Shixia Liu, Haixun Wang
        In Proceedings of the SIGKDD Conference on Knowledge Discovery and Data Mining (17.6%) (KDD 2012),
        Beijing, China, August 2012.


    • [Others]

      • Visualizing Path Exploration to Assist Problem Diagnosis for Structural Test Generation
        Jiayi Cao, Angello Astorga, Siwakorn Srisakaokul, Zhengkai Wu, Xueqing Liu, Xusheng Xiao, and Tao Xie.
        In Proceedings of the IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC 2018), Posters, Lisbon, Portugal, October 2018.
        [Short Paper], [Slides]


      Kudos to Vasilios for the template: Plain Academic