Reporting computing projects through structured abstracts: a quasi-experiment

Previous work has demonstrated that the use of structured abstracts can lead to greater completeness and clarity of information, making it easier for researchers to extract information about a study. In academic year 2007/08, Durham University’s Computer Science Department revised the format of the project report that final year students were required to write, from a ‘traditional dissertation’ format, using a conventional abstract, to that of a 20-page technical paper, together with a structured abstract. This study set out to determine whether inexperienced authors (students writing their final project reports for computing topics) find it easier to produce good abstracts, in terms of completeness and clarity, when using a structured form rather than a conventional form. We performed a controlled quasi-experiment in which a set of ‘judges’ each assessed one conventional and one structured abstract for its completeness and clarity. These abstracts were drawn from those produced by four cohorts of final year students: two preceding the change, and the two following. The assessments were performed using a form of checklist that is similar to those used for previous experimental studies. We used 40 abstracts (10 per cohort) and 20 student ‘judges’ to perform the evaluation. Scored on a scale of 0.1–1.0, the mean for completeness increased from 0.37 to 0.61 when using a structured form. For clarity, using a scale of 1–10, the mean score increased from 5.1 to 7.2. For a minimum goal of scoring 50% for both completeness and clarity, only 3 from 19 conventional abstracts achieved this level, while only 3 from 20 structured abstracts failed to reach it. We conclude that the use of a structured form for organising the material of an abstract can assist inexperienced authors with writing technical abstracts that are clearer and more complete than those produced without the framework provided by such a mechanism.     

The search for a research method for studying OSS process innovation

Medium-sized, open-participation Open Source Software (OSS) projects do not usually perform explicit software process improvement on any routine basis. It would be useful to understand how to get such a project to accept a process improvement proposal and hence to perform process innovation. We want to determine an effective and feasible qualitative research method for studying the above question. We present (narratively) a case study of how we worked towards and eventually found such a research method. The case involves four attempts at collecting suitable data about innovation episodes (direct participation (twice), polling developers for episodes, manually finding episodes in mailing list archives) and the adaptation of the Grounded Theory data analysis methodology. Direct participation allows gathering rather rich data, but does not allow for observing a sufficiently large number of innovation episodes. Polling developers for episodes did not prove to be useful. Using mailing list archives to find data to be analyzed is both feasible and effective. We also describe how the data thus found can be analyzed based on the Grounded Theory Method with suitable adjustments. By-and-large, our findings ought to apply to studying various phenomena in OSS development processes that are similarly heavyweight and infrequent. However, specific details may block this possibility and we cannot predict which details that might be. The amount of effort involved in direct participation approaches to qualitative research can easily be underestimated. Also, survey approaches are not well-suited for many process issues in OSS, because too few developers are sufficiently process-conscious. An approach based on passive observation is a viable alternative in the OSS context due to the availability of large amounts of fairly complete archival data.     

Dynamics of a large scale rigid–flexible multibody system composed of composite laminated plates

A novel computational approach for the dynamic analysis of a large scale rigid–flexible multibody system composed of composite laminated plates is proposed. The rigid parts in the system are described through the Natural Coordinate Formulation (NCF) and the flexible bodies in the system are modeled via the finite elements of Absolute Nodal Coordinate Formulation (ANCF), which can lead to a constant mass matrix for the derived system equation of motion. For modeling composite laminated plates accurately, a new composite laminated plate element of ANCF is proposed and the corresponding efficient formulations for evaluating both the elastic force and its Jacobian of the element are derived from the first Piola–Kirchhoff stress tensor. To improve computational efficiency, the sparse matrix technology and graph theory are used to solve the huge set of linear algebraic equations in the process of integrating the equations of motion by using the generalized-a method, and an OpenMP based parallel scheme is also introduced. Finally, the effectiveness of the proposed approach is validated through two numerical examples. One is the static simulation of a single composite laminated plate under gravity and the other is the dynamic simulations of unfolding process of a satellite system with a pair of complicated antennas.     

Adaptive control of straight worms without derivative measurement

In this paper we consider an adaptive control problem of finite DoF worm-like locomotion systems (WLLS) which contact the ground with Coulomb dry friction. Using a rough mathematical friction law the system is shown to belong to a system class that allows adaptive control. Gaits from the kinematic theory can be tracked by means of adaptive controllers. For this we introduce two different adaptive controllers for λ-tracking and focus on that one which is not based on the derivative of the output. We pay attention to the analysis of such systems and present some theoretical control investigations including proofs. Numerical simulations of tracking different reference signals under arbitrary choice of the system parameters demonstrate and illustrate that the introduced simple adaptive controllers work successfully and effectively. Current experiments are aimed at the justification of theoretical results.     

Manifold elastic net: a unified framework for sparse dimension reduction

It is difficult to find the optimal sparse solution of a manifold learning based dimensionality reduction algorithm. The lasso or the elastic net penalized manifold learning based dimensionality reduction is not directly a lasso penalized least square problem and thus the least angle regression (LARS) (Efron et al., Ann Stat 32(2):407–499, 2004), one of the most popular algorithms in sparse learning, cannot be applied. Therefore, most current approaches take indirect ways or have strict settings, which can be inconvenient for applications. In this paper, we proposed the manifold elastic net or MEN for short. MEN incorporates the merits of both the manifold learning based dimensionality reduction and the sparse learning based dimensionality reduction. By using a series of equivalent transformations, we show MEN is equivalent to the lasso penalized least square problem and thus LARS is adopted to obtain the optimal sparse solution of MEN. In particular, MEN has the following advantages for subsequent classification: (1) the local geometry of samples is well preserved for low dimensional data representation, (2) both the margin maximization and the classification error minimization are considered for sparse projection calculation, (3) the projection matrix of MEN improves the parsimony in computation, (4) the elastic net penalty reduces the over-fitting problem, and (5) the projection matrix of MEN can be interpreted psychologically and physiologically. Experimental evidence on face recognition over various popular datasets suggests that MEN is superior to top level dimensionality reduction algorithms.     

The adoption and concerns of e-finance in Malaysia

Technological advancement has brought a whole new way of delivering financial services around the world. E-finance industry in Malaysia has grown rapidly with the help of internet and conducting financial services are far easier now if compared to last time. Internet provided a new medium or channel in delivering financial services. However, there are also risks and challenges when using electronic finance. A total of 100 respondents had been targeted in completing this study. Survey method was utilized in collecting respondent’s opinion and perception to test the variables. The target was companies which used electronic finance in delivering their services. Interviews were also conducted and relevant tests conducted which provided evidence to support the hypotheses for this study. It was found that e-finance was affected by security, revenue and cost dimensions, and technology architecture with the adoption of global technology.     

On improved calibration method for the catadioptric omnidirectional vision with a single viewpoint

The single viewpoint constraint is a principal optical characteristic for most catadioptric omnidirectional vision. Single viewpoint catadioptric omnidirectional vision is very useful because it allows the generation of geometrically correct perspective images from one omnidirectional image. Therefore precise calibration for single viewpoint constraint is needed during system assembling. However, in most image detection based calibration methods, the nonlinear optical distortion brought by lens is often neglected. Hence the calibration precision is poor. In this paper, a new calibration method of single viewpoint constraint for the catadioptric omni-directional vision is proposed. Firstly, an image correction algorithm is obtained by training a neural network. Then, according to characteristics of the space circular perspective projection, the corrected image of the mirror boundary is used to estimate its position and attitude relative to the camera to guide the calibration. Since the estimate is conducted based on actual imaging model rather than the simplified model, the estimate error is largely reduced, and the calibration accuracy is significantly improved. Experiments are conducted on simulated images and real images to show the accuracy and the effectiveness of the proposed method.