Genomics data is essential in biological and medical domains, and bioinformatics analysts often manually create circos plots to analyze the data and extract valuable insights. However, creating circos plots is complex, as it requires careful design for multiple track attributes and positional relationships between them. Typically, analysts often seek inspiration from existing circos plots, and they have to iteratively adjust and refine the plot to achieve a satisfactory final design, making the process both tedious and time-intensive. To address these challenges, we propose IntelliCircos, an AI-powered interactive authoring tool that streamlines the process from initial visual design to the final implementation of circos plots. Specifically, we build a new dataset containing 4396 circos plots with corresponding annotations and configurations, which are extracted and labeled from published papers. With the dataset, we further identify track combination patterns, and utilize Large Language Model (LLM) to provide domain-specific design recommendations and configuration references to navigate the design of circos plots. We conduct a user study with 8 bioinformatics analysts to evaluate IntelliCircos, and the results demonstrate its usability and effectiveness in authoring circos plots.

Computational notebooks are widely utilized for exploration and analysis. However, creating slides to communicate analysis results from these notebooks is quite tedious and time-consuming. Researchers have proposed automatic systems for generating slides from notebooks, which, however, often do not consider the process of users conceiving and organizing their messages from massive code cells. Those systems ask users to go directly into the slide creation process, which causes potentially ill-structured slides and burdens in further refinement. Inspired by the common and widely recommended slide creation practice: drafting outlines first and then adding concrete content, we introduce OutlineSpark, an AI-powered slide creation tool that generates slides from a slide outline written by the user. The tool automatically retrieves relevant notebook cells based on the outlines and converts them into slide content. We evaluated OutlineSpark with 12 users. Both the quantitative and qualitative feedback from the participants verify its effectiveness and usability.

Congenital heart disease (CHD) is the most common congenital malformation and imaging examination is an important means to diagnose it. Currently, deep learning-based methods have achieved remarkable results in various types of imaging examinations. However, the issues of large parameter size and low throughput limit their clinical applications. In this paper, we design an efficient, light-weight hybrid model named FlashViT, to assist cardiovascular radiologists in early screening and diagnosis of CHD. Specifically, we propose the Large-scale Token Merging Module (LTM) for more aggressive similar token merging without sacrificing accuracy, which alleviate the problem of high computational complexity and resource consumption of self-attention mechanism. In addition, we propose an unsupervised homogenous pre-training strategy to tackle the issue of insufficient medical image data and poor generalization ability. Compared with conventional pre-training strategy that use ImageNet1K, our strategy only utilizes less than 1% of the class-agnostic medical images from ImageNet1K, resulting in faster convergence speed and advanced performance of the model. We conduct extensive validation on the collected CHD dataset and the results indicate that our proposed FlashViT-S achieves accuracy of 92.2% and throughput of 3753 fps with about 3.8 million parameters. We hope that this work can provide some assistance in designing laboratory models for future application in clinical practice.

Feature selection contributes to increasing many learners' accuracy by identifying and removing irrelevant features in multidimensional datasets. Conventional feature selection methods determine the optimal feature subset independently from and prior to the introduction of a new query. In general, some features will be relevant only in certain tasks. We argue that a query, as an indicator of the attention focus and current task, is a major part of the context and should be involved in the determination of the final feature subset. In this paper we attempt to propose a query-sensitive feature selection model, present two algorithms for applying such a feature selection method, and test their effectiveness by comparing their performances to those of the conventional selection algorithms. Our experiments are executed under a nearest neighbor classification environment and the results show a consistent improvement in the classification performance when a query-sensitive feature subset is selected and used for measuring similarities between the query and other instances. The results suggest that the performance of a lazy learner has the potential to increase through query-sensitive feature selection

Case based reasoning (CBR) is a problem solving method that reuses the previous problem solving experiences (represented as cases) to solve the new problem. As a type of interactive CBR, conversational CBR has been proposed to help users construct their problem descriptions incrementally through a mixed-initiative question-answering sequence. In software component retrieval, users meet the difficulty in well-defining their component queries. As a solution to release this difficulty, we proposed and implemented a conversational component retrieval model (CCRM) using a knowledge-intensive conversational CBR method in the thesis.The research activities and contributions followed two directions: theoretical research on conversational CBR to provide an efficient and natural conversation process, and applying and adapting conversational CBR to support software component retrieval.In the theoretical research direction, we first provided a framework to classify the similarity methods in CBR from the perspective of what features were taken into account during the similarity calculation, and further analyzed and illustrated that the query-biased similarity methods (only considering the features appearing in the query during the similarity calculation) were more suitable for conversational CBR applications. A knowledge-intensive conversational CBR method was designed that was able to utilize the general domain knowledge to improve the efficiency and naturalness of the conversation process. Four knowledge-intensive question selection tasks, including the feature inferencing, the integrated question ranking, the consistent question clustering, and the coherent question sequencing, were identified and handled in this method. We also proposed a lazy dialog learning mechanism that could continuously improve the performance of conversational CBR.Following the conversational component retrieval application direction, we reviewed and analyzed the current software component retrieval methods and proposed a conversational component retrieval model. In order to represent both the software components and the component queries as cases, it is necessary for a case to have multiple values on some features (generalized cases). In the research, we analyzed the feasibilities and discussed the methods to extend conversational CBR to support generalized cases from three aspects: the case representation, the similarity calculation metric, and the question selection method. At the end, a knowledge-intensive conversational software component retrieval system (TrollCCRM), enhanced by the above research findings, was implemented and evaluated on the image processing software component retrieval application.The evaluation results so far gave us positive results that the TrollCCRM system provided an efficient and natural conversation process guiding users to find their desired software components.

Case based reasoning (CBR) is a problem solving method that reuses the previous problem solving experiences (represented as cases) to solve the new problem. As a type of interactive CBR, conversational CBR has been proposed to help users construct their problem descriptions incrementally through a mixed-initiative question-answering sequence. In software component retrieval, users meet the difficulty in well-defining their component queries. As a solution to release this difficulty, we proposed and implemented a conversational component retrieval model (CCRM) using a knowledge-intensive conversational CBR method in the thesis.The research activities and contributions followed two directions: theoretical research on conversational CBR to provide an efficient and natural conversation process, and applying and adapting conversational CBR to support software component retrieval.In the theoretical research direction, we first provided a framework to classify the similarity methods in CBR from the perspective of what features were taken into account during the similarity calculation, and further analyzed and illustrated that the query-biased similarity methods (only considering the features appearing in the query during the similarity calculation) were more suitable for conversational CBR applications. A knowledge-intensive conversational CBR method was designed that was able to utilize the general domain knowledge to improve the efficiency and naturalness of the conversation process. Four knowledge-intensive question selection tasks, including the feature inferencing, the integrated question ranking, the consistent question clustering, and the coherent question sequencing, were identified and handled in this method. We also proposed a lazy dialog learning mechanism that could continuously improve the performance of conversational CBR.Following the conversational component retrieval application direction, we reviewed and analyzed the current software component retrieval methods and proposed a conversational component retrieval model. In order to represent both the software components and the component queries as cases, it is necessary for a case to have multiple values on some features (generalized cases). In the research, we analyzed the feasibilities and discussed the methods to extend conversational CBR to support generalized cases from three aspects: the case representation, the similarity calculation metric, and the question selection method. At the end, a knowledge-intensive conversational software component retrieval system (TrollCCRM), enhanced by the above research findings, was implemented and evaluated on the image processing software component retrieval application.The evaluation results so far gave us positive results that the TrollCCRM system provided an efficient and natural conversation process guiding users to find their desired software components.

One difficulty in software component retrieval comes from users’ incapability to well define their queries. In this paper, we propose a conversational component retrieval model (CCRM) to alleviate this difficulty. CCRM uses a knowledge-intensive conversational case-based reasoning method to help users to construct their queries incrementally through a mixed-initiative question-answering process. In this model, general domain knowledge is captured and utilized in helping tackle the following five tasks: feature inferencing, semantic similarity calculation, integrated question ranking, consistent question clustering and coherent question sequencing. This model is implemented, and evaluated in an image processing component retrieval application. The evaluation result gives us positive support.

The complexity and high construction cost of case bases make it very difficult, if not impossible, to evaluate a CBR system, especially a knowledge-intensive CBR system, using statistical evaluation methods on many case bases. In this paper, we propose an evaluation strategy, which uses both many simple case bases and a few complex case bases to evaluate a CBR system, and show how this strategy may satisfy different evaluation goals. The identified evaluation goals are classified into two categories: domain-independent and domain-dependent. For the evaluation goals in the first category, we apply the statistical evaluation method using many simple case bases (for example, UCI data sets); for evaluation goals in the second category, we apply different, relatively weak, evaluation methods on a few complex domain-specific case bases. We apply this combined evaluation strategy to evaluate our knowledge-intensive conversational CBR method as a case study.

In conversational case-based reasoning (CCBR), a main problem is how to select the most discriminative questions and display them to users in a natural way to alleviate users’ cognitive load. This is referred to as the question selection task. Current question selection methods are knowledge-poor, that is, only statistical metrics are taken into account. In this paper, we identify four computational tasks of a conversation process: feature inferencing, question ranking, consistent question clustering and coherent question sequencing. We show how general domain knowledge is able to improve these processes. A knowledge representation system suitable for capturing both cases and general knowledge has been extended with meta-level relations for controlling a CCBR process. An “explanation-boosted” reasoning approach, designed to accomplish the knowledge-intensive question selection tasks, is presented. An application of our implemented system is illustrated in the car fault detection domain.

Conversational Case-Based-Reasoning (CCBR) provides a mixed-initiative dialog for guiding users to construct their problem description incrementally through a question-answering sequence. Similarity calculation in CCBR, as in traditional CBR, plays an important role in the retrieval process since it decides the quality of the retrieved case. In this paper, we analyze the different characteristics of the query (new case) between CCBR and traditional CBR, and argue that the similarity calculation method that only takes the features appearing in the query into account, so called query-biased, is more suitable for CCBR. An experiment is designed and executed on 36 datasets. The results show us that on 31 datasets out of the total 36, the CCBR system using the query-biased similarity calculation method achieves more effective performance than those using case-biased and equally-biased similarity calculation methods.

Component retrieval, about how to locate and identify appropriate components, is one of the major problems in component reuse. It becomes more critical as more reusable components come from component markets instead of from an in-house component library, and the number of available components is dramatically increasing. In this paper, we review the current component retrieval methods and propose our conversational component retrieval model (CCRM). In CCRM, components are represented as cases, a knowledge-intensive case-based reasoning (CBR) method is adopted to explore context-based semantic similarities between users' query and stored components, and a conversational case-based reasoning (CCBR) technology is selected to acquire users' requirements interactively and incrementally.

Conversational Case-Based Reasoning (CCBR) provides a mixed-initiative dialog for guiding users to refine their problem descriptions incrementally through a question-answering sequence. Most CCBR approaches assume that there is at most one discrete value on each feature. While a generalized case (GC), which has been proposed and used in traditional CBR processes, has multiple values on some features. Motivated by the conversational software component retrieval application, we focus on the problem of extending CCBR to support GCs in this paper. This problem is tackled from two aspects: similarity measuring and discriminative question ranking.

One difficulty in component retrieval comes from users' incapability to well define their queries. In this paper, we propose a conversational component retrieval model (CCRM) to alleviate this difficulty. In CCRM, a knowledge-intensive conversational case-based reasoning method is adopted to infer potential knowledge from current known knowledge, calculate the context-based semantic similarities between users' queries and stored components, and prompt users the most discriminative questions to extract more information to refine their component queries interactively and incrementally.

Though more and more researchers have realized the importance of creativity in software development, there are few empirical studies reported on this topic. In this paper, we present an exploratory empirical research in which several issues on creativity in software development are studied, that is, which development phases are perceived to include more creative work, whether or not UML-based documentation can make developers perceive more time is devoted to creative work, whether or not more creative work can accelerate the software development speed, and whether developers more prefer to do the creative work. Based on result analyses, we proposed four hypotheses to direct the future research in this field and discussed the challenge that 'since developers do not like to participate in those improving activities (quality assuring activities), how can we keep and improve software quality effectively and efficiently?'.

A core research concern in conversational case-based reasoning (CCBR) is how to select the most discriminative and natural questions to ask the user in the conversational process. There are two ways to realize this task: one is to remove the questions whose answers can be inferred from the information a user has provided, which is called dialogue inferencing; the other is to rank the questions to guarantee the most informative questions are asked first, which is referred to as question ranking. In this paper, we present a common explanation-boosted CCBR approach, which utilizes both general domain knowledge and case-specific knowledge to realize dialogue inferencing and question ranking. This approach provides a flexible meta-level knowledge representation model to be able to incorporate richer semantic relations. An application of this approach is illustrated in a car fault detection domain.

One major problem of nearest neighbor (NN) algorithms is inefficiency incurred by irrelevant features. A solution to this problem is to assign weights to features that indicate their salience for classification. Current weighting methods can be divided as global weighting, partial local weighting, and local weighting methods enumerated in increasing order of capability to capture the features' relative salience in classification. However, the existing methods are not sensitive enough to describe the salience of a feature and can be changed given different queries. We suggest that the salience of a feature, in addition to being sensitive to the instance (i.e. varies across instances), should also be sensitive to the variations in the difference of a feature's values between a query and the instances in the instance base. In this paper, we put forward a dynamic feature weighting approach which has more expressive capability, and present a sketch of a classification algorithm based on the notion of dynamic weights.

Case based reasoning (CBR) is a problem solving method that reuses the previous problem solving experiences (represented as cases) to solve the new problem. As a type of interactive CBR, conversational CBR has been proposed to help users construct their problem descriptions incrementally through a mixed-initiative question-answering sequence. In software component retrieval, users meet the difficulty in well-defining their component queries. As a solution to release this difficulty, we proposed and implemented a conversational component retrieval model (CCRM) using a knowledge-intensive conversational CBR method in the thesis.The research activities and contributions followed two directions: theoretical research on conversational CBR to provide an efficient and natural conversation process, and applying and adapting conversational CBR to support software component retrieval.In the theoretical research direction, we first provided a framework to classify the similarity methods in CBR from the perspective of what features were taken into account during the similarity calculation, and further analyzed and illustrated that the query-biased similarity methods (only considering the features appearing in the query during the similarity calculation) were more suitable for conversational CBR applications. A knowledge-intensive conversational CBR method was designed that was able to utilize the general domain knowledge to improve the efficiency and naturalness of the conversation process. Four knowledge-intensive question selection tasks, including the feature inferencing, the integrated question ranking, the consistent question clustering, and the coherent question sequencing, were identified and handled in this method. We also proposed a lazy dialog learning mechanism that could continuously improve the performance of conversational CBR.Following the conversational component retrieval application direction, we reviewed and analyzed the current software component retrieval methods and proposed a conversational component retrieval model. In order to represent both the software components and the component queries as cases, it is necessary for a case to have multiple values on some features (generalized cases). In the research, we analyzed the feasibilities and discussed the methods to extend conversational CBR to support generalized cases from three aspects: the case representation, the similarity calculation metric, and the question selection method. At the end, a knowledge-intensive conversational software component retrieval system (TrollCCRM), enhanced by the above research findings, was implemented and evaluated on the image processing software component retrieval application.The evaluation results so far gave us positive results that the TrollCCRM system provided an efficient and natural conversation process guiding users to find their desired software components.

Nowadays, many empirical studies are carried out based on educational projects. In these cases, instructors put forward requirements on educational projects from the perspective of education, and researchers also provide their requirements from the perspective of empirical study. How to design educational projects successfully from both educational perspective and research perspective becomes a practical challenge. In this paper, we present a successful case to resolve this problem: we designed two educational projects in a software architecture course, based on which an empirical study about creativity in software development was carried out. Our methods to design these educational projects can be reused for other software engineering courses associated with empirical study.