arXiv subjects

Liming Zhu

Publications and source records attributed to Liming Zhu.

139 records · Page 8Linked to original sources

arXiv2020

Task-Oriented API Usage Examples Prompting Powered By Programming Task Knowledge Graph

Programming tutorials are often created to demonstrate programming tasks with code examples. However, our study of Stack Overflow questions reveals the low utilization of high-quality programming tutorials, which is caused task description mismatch and code information overload. Document search can find relevant tutorial documents, but they often cannot find specific programming actions and code solutions relevant to the developers' task needs. The recently proposed activity-centric search over knowledge graph supports direct search of programming actions, but it has limitations in action coverage, natural language based task search, and coarse-grained code example recommendation. In this work, we enhance action coverage in knowledge graph with actions extracted from comments in code examples and more forms of activity sentences. To overcome the task description mismatch problem, we develop a code matching based task search method to find relevant programming actions and code examples to the code under development. We integrate our knowledge graph and task search method in the IDE, and develop an observe-push based tool to prompt developers with task-oriented API usage examples. To alleviate the code information overload problem, our tool highlights programming action and API information in the prompted tutorial task excerpts and code examples based on the underlying knowledge graph. Our evaluation confirms the high quality of the constructed knowledge graph, and show that our code matching based task search can recommend effective code solutions to programming issues asked on Stack Overflow. A small-scale user study demonstrates that our tool is useful for assisting developers in finding and using relevant programming tutorials in their programming tasks.

cs.SE↗

arXiv2020

Design-Pattern-as-a-Service for Blockchain-based Self-Sovereign Identity

Self-sovereign identity (SSI) is considered to be a "killer application" of blockchain. However, there is a lack of systematic architecture designs for blockchain-based SSI systems to support methodical development. An aspect of such gap is demonstrated in current solutions, which are considered coarse grained and may increase data security risks. In this paper, we first identify the lifecycles of three major SSI objects (i.e., key, identifier, and credential) and present fine-grained design patterns critical for application development. These patterns are associated with particular state transitions, providing a systematic view of system interactions and serving as a guidance for effective use of these patterns. Further, we present an SSI platform architecture, which advocates the notion of Design-Pattern-as-a-Service. Each design pattern serves as an API by wrapping the respective pattern code to ease application development and improve scalability and security. We implement a prototype and evaluate it on feasibility and scalability.

cs.SE↗

arXiv2019

uBaaS: A Unified Blockchain as a Service Platform

Blockchain is an innovative distributed ledger technology which has attracted a wide range of interests for building the next generation of applications to address lack-of-trust issues in business. Blockchain as a service (BaaS) is a promising solution to improve the productivity of blockchain application development. However, existing BaaS deployment solutions are mostly vendor-locked: they are either bound to a cloud provider or a blockchain platform. In addition to deployment, design and implementation of blockchain-based applications is a hard task requiring deep expertise. Therefore, this paper presents a unified blockchain as a service platform (uBaaS) to support both design and deployment of blockchain-based applications. The services in uBaaS include deployment as a service, design pattern as a service and auxiliary services. In uBaaS, deployment as a service is platform agnostic, which can avoid lock-in to specific cloud platforms, while design pattern as a service applies design patterns for data management and smart contract design to address the scalability and security issues of blockchain. The proposed solutions are evaluated using a real-world quality tracing use case in terms of feasibility and scalability.

cs.SE↗

arXiv2019

Adversarial Examples on Graph Data: Deep Insights into Attack and Defense

Graph deep learning models, such as graph convolutional networks (GCN) achieve remarkable performance for tasks on graph data. Similar to other types of deep models, graph deep learning models often suffer from adversarial attacks. However, compared with non-graph data, the discrete features, graph connections and different definitions of imperceptible perturbations bring unique challenges and opportunities for the adversarial attacks and defenses for graph data. In this paper, we propose both attack and defense techniques. For attack, we show that the discreteness problem could easily be resolved by introducing integrated gradients which could accurately reflect the effect of perturbing certain features or edges while still benefiting from the parallel computations. For defense, we observe that the adversarially manipulated graph for the targeted attack differs from normal graphs statistically. Based on this observation, we propose a defense approach which inspects the graph and recovers the potential adversarial perturbations. Our experiments on a number of datasets show the effectiveness of the proposed methods.

cs.LG↗

arXiv2018

An Empirical Study of Architecting for Continuous Delivery and Deployment

Recently, many software organizations have been adopting Continuous Delivery and Continuous Deployment (CD) practices to develop and deliver quality software more frequently and reliably. Whilst an increasing amount of the literature covers different aspects of CD, little is known about the role of software architecture in CD and how an application should be (re-) architected to enable and support CD. We have conducted a mixed-methods empirical study that collected data through in-depth, semi-structured interviews with 21 industrial practitioners from 19 organizations, and a survey of 91 professional software practitioners. Based on a systematic and rigorous analysis of the gathered qualitative and quantitative data, we present a conceptual framework to support the process of (re-) architecting for CD. We provide evidence-based insights about practicing CD within monolithic systems and characterize the principle of "small and independent deployment units" as an alternative to the monoliths. Our framework supplements the architecting process in a CD context through introducing the quality attributes (e.g., resilience) that require more attention and demonstrating the strategies (e.g., prioritizing operations concerns) to design operations-friendly architectures. We discuss the key insights (e.g., monoliths and CD are not intrinsically oxymoronic) gained from our study and draw implications for research and practice.

cs.SE↗

arXiv2018

Metric Factorization: Recommendation beyond Matrix Factorization

In the past decade, matrix factorization has been extensively researched and has become one of the most popular techniques for personalized recommendations. Nevertheless, the dot product adopted in matrix factorization based recommender models does not satisfy the inequality property, which may limit their expressiveness and lead to sub-optimal solutions. To overcome this problem, we propose a novel recommender technique dubbed as {\em Metric Factorization}. We assume that users and items can be placed in a low dimensional space and their explicit closeness can be measured using Euclidean distance which satisfies the inequality property. To demonstrate its effectiveness, we further designed two variants of metric factorization with one for rating estimation and the other for personalized item ranking. Extensive experiments on a number of real-world datasets show that our approach outperforms existing state-of-the-art by a large margin on both rating prediction and item ranking tasks.

cs.IR↗

arXiv2018

Blockchains for Business Process Management - Challenges and Opportunities

Blockchain technology promises a sizable potential for executing inter-organizational business processes without requiring a central party serving as a single point of trust (and failure). This paper analyzes its impact on business process management (BPM). We structure the discussion using two BPM frameworks, namely the six BPM core capabilities and the BPM lifecycle. This paper provides research directions for investigating the application of blockchain technology to BPM.

cs.SE↗

arXiv2017

Interpreting Shared Deep Learning Models via Explicable Boundary Trees

Despite outperforming the human in many tasks, deep neural network models are also criticized for the lack of transparency and interpretability in decision making. The opaqueness results in uncertainty and low confidence when deploying such a model in model sharing scenarios, when the model is developed by a third party. For a supervised machine learning model, sharing training process including training data provides an effective way to gain trust and to better understand model predictions. However, it is not always possible to share all training data due to privacy and policy constraints. In this paper, we propose a method to disclose a small set of training data that is just sufficient for users to get the insight of a complicated model. The method constructs a boundary tree using selected training data and the tree is able to approximate the complicated model with high fidelity. We show that traversing data points in the tree gives users significantly better understanding of the model and paves the way for trustworthy model sharing.

cs.LG↗

arXiv2017

Hybrid Collaborative Recommendation via Semi-AutoEncoder

In this paper, we present a novel structure, Semi-AutoEncoder, based on AutoEncoder. We generalize it into a hybrid collaborative filtering model for rating prediction as well as personalized top-n recommendations. Experimental results on two real-world datasets demonstrate its state-of-the-art performances.

cs.IR↗

arXiv2017

Design Choices for Data Governance in Platform Ecosystems: A Contingency Model

As platform ecosystems are growing by platform users' data, the importance of data governance has been highlighted. In particular, how to share control and decision rights with platform users are regarded as significant design issues since the role of them is increasing. Platform context should be considered when designing data governance in platform ecosystems (i.e. centralized/decentralized governance). However, there is limited research on this issue. Existing models focus on characteristics for enterprises. This results in limited support for platform ecosystems where there are different types of business context such as open strategies or platform maturity. This paper develops a contingency model for platform ecosystems including distinctive contingency factors. The study then discusses which data governance factors should be carefully considered and strengthened for each contingency in order to succeed in governance and to win market. A case study is performed to validate our model and to show its practical implications.

cs.SE↗

arXiv2017

Data Governance for Platform Ecosystems: Critical Factors and the State of Practice

Recently, platform ecosystem has received attention as a key business concept. Sustainable growth of platform ecosystems is enabled by platform users supplying and/or demanding content from each other: e.g. Facebook, YouTube or Twitter. The importance and value of user data in platform ecosystems is accentuated since platform owners use and sell the data for their business. Serious concern is increasing about data misuse or abuse, privacy issues and revenue sharing between the different stakeholders. Traditional data governance focuses on generic goals and a universal approach to manage the data of an enterprise. It entails limited support for the complicated situation and relationship of a platform ecosystem where multiple participating parties contribute, use data and share profits. This article identifies data governance factors for platform ecosystems through literature review. The study then surveys the data governance state of practice of four platform ecosystems: Facebook, YouTube, EBay and Uber. Finally, 19 governance models in industry and academia are compared against our identified data governance factors for platform ecosystems to reveal the gaps and limitations.

cs.CY↗

arXiv2017

HPDedup: A Hybrid Prioritized Data Deduplication Mechanism for Primary Storage in the Cloud

Eliminating duplicate data in primary storage of clouds increases the cost-efficiency of cloud service providers as well as reduces the cost of users for using cloud services. Existing primary deduplication techniques either use inline caching to exploit locality in primary workloads or use post-processing deduplication running in system idle time to avoid the negative impact on I/O performance. However, neither of them works well in the cloud servers running multiple services or applications for the following two reasons: Firstly, the temporal locality of duplicate data writes may not exist in some primary storage workloads thus inline caching often fails to achieve good deduplication ratio. Secondly, the post-processing deduplication allows duplicate data to be written into disks, therefore does not provide the benefit of I/O deduplication and requires high peak storage capacity. This paper presents HPDedup, a Hybrid Prioritized data Deduplication mechanism to deal with the storage system shared by applications running in co-located virtual machines or containers by fusing an inline and a post-processing process for exact deduplication. In the inline deduplication phase, HPDedup gives a fingerprint caching mechanism that estimates the temporal locality of duplicates in data streams from different VMs or applications and prioritizes the cache allocation for these streams based on the estimation. HPDedup also allows different deduplication threshold for streams based on their spatial locality to reduce the disk fragmentation. The post-processing phase removes duplicates whose fingerprints are not able to be cached due to the weak temporal locality from disks. Our experimental results show that HPDedup clearly outperforms the state-of-the-art primary storage deduplication techniques in terms of inline cache efficiency and primary deduplication efficiency.

cs.DC↗

arXiv2017

Continuous Integration, Delivery and Deployment: A Systematic Review on Approaches, Tools, Challenges and Practices

Context: Continuous practices, i.e., continuous integration, delivery, and deployment, are the software development industry practices that enable organizations to frequently and reliably release new features and products. With the increasing interest in and literature on continuous practices, it is important to systematically review and synthesize the approaches, tools, challenges, and practices reported for adopting and implementing continuous practices. Objective: This research aimed at systematically reviewing the state of the art of continuous practices to classify approaches and tools, identify challenges and practices in this regard, and identify the gaps for future research. Method: We used systematic literature review (SLR) method for reviewing the peer-reviewed papers on continuous practices published between 2004 and 1st June 2016. We applied thematic analysis method for analysing the data extracted from reviewing 69 papers selected using predefined criteria. Results: We have identified thirty approaches and associated tools, which facilitate the implementation of continuous practices in the following ways: (1) "reducing build and test time in continuous integration (CI)"; (2) "increasing visibility and awareness on build and test results in CI"; (3) "supporting (semi-) automated continuous testing"; (4) "detecting violations, flaws and faults in CI"; (5) "addressing security and scalability issues in deployment pipeline", and (6) "improving dependability and reliability of deployment process". We have also determined a list of critical factors such as "testing (effort and time)", "team awareness and transparency", "good design principles", "customer", "highly skilled and motivated team", "application domain", and "appropriate infrastructure" that should be carefully considered when introducing continuous practices in a given organization.

cs.SE↗