Maxim Bakaev

A quantitative measure for information transfer in human-machine control systems

This study is an exploration in measuring quantities of information transferred in the course of interaction in human-machine systems, which deemed necessary as fundamental support for interface design. We put forwards several possible measures found in previous research works, including the infamous Hicks law, and validate their applicability with dedicated experiment with 33 users performing selection tasks. The results suggest that interface “effective information capacity” factor, incorporating number of alternative targets (N) and logarithm of their vocabulary size (K, possible different symbols), provided better fit to the observed performance time, while Hicks law model and “interface message length” (IML) based models were clearly inferior. ANOVA suggested strong effect of interaction between N, K, and IML, but building corresponding conclusive model would require further theoretical research. We also calculated selection tasks throughput, which amounted to 59.0 bit/s in the experiment, steadily increasing per K and decreasing with age. Our findings might aid interface designers, in particular ones developing human-machine control systems that seek to maximize overall operating efficiency and minimize costly human errors.

Measuring and Ensuring Similarity of User Interfaces: The Impact of Web Layout

Given the rapid update cycles in modern web information systems and the abundance of legacy software being migrated to the web, controlling similarity between user interfaces (UI) is an actual problem of interaction engineering. The similarity (consistency) aspect is also increasingly considered in computer-aided design, where it is included in the optimized goal function, to minimize re-learning effort for users. In this paper, we explore the impact of the proposed layout distance measure, which is calculated for different levels of hierarchy in web UIs, which we identify as: Region – Block – Group – Element. To support our approach, we conducted an experimental pilot study in the context of an ongoing medical information system (IS) web migration project. The regression analysis suggests that layout distance (particularly, its orientation dimension) does have effect on web UI similarity as perceived by users. The results can be used by web engineers, in particular to smoothen the transition between versions of a UI for users and IS operators.

A case-based reasoning approach with fuzzy linguistic rules: Accuracy validation and application in interface design-support intelligent system

The paper researches novel approach towards estimating relevance between cases in case-based reasoning system. Fuzzy linguistic rules are used as a tool for representing knowledge about cases relevance. The results suggest that the proposed hybrid system employing case model and fuzzy logic can work with even a small number of cases, while yielding accurate results. We also provide an example of the model application in practice — in a dedicated design-support intelligent system. The solutions created with the system demonstrated reasonable quality in terms of usability and, to some extent, in subjective attractiveness for target users.

Web Intelligence Linked Open Data for Website Design Reuse

Code and design reuse are as old as software engineering industry itself, but it’s also always a new trend, as more and more software products and websites are being created. Domain-specific design reuse on the web has especially high potential, saving work effort for thousands of developers and encouraging better interaction quality for millions of Internet users. In our paper we perform pilot feature engineering for finding similar solutions (website designs) within Domain, Task, and User UI models supplemented by Quality aspects. To obtain the feature values, we propose extraction of website-relevant data from online global services (DMOZ, Alexa, SimilarWeb, etc.) considered as linked open data sources, using specially developed web intelligence data miner. The preliminary investigation with 21 websites and 82 human annotators showed reasonable accuracy of the data sources and suggests potential feasibility of the approach.

Extending Kansei Engineering for Requirements Consideration in Web Interaction Design

In our paper we consider how the eminent Kansei Engineering (KE) method can be applied in computer-aided development of websites. Although principally used for exploring emotional dimension of users’ experience with products, KE can be extended to incorporate other types of software requirements. In conjunction with AI Neural Networks (Kansei Type II), it then becomes possible to automate, up to a certain degree, evaluation of website quality in terms of functionality, usability, and appeal. We provide an overview of existing works related to KE application in web design, and note its certain gap with systematic Web Engineering. Then we summarize approaches for auto-validation of different types of requirements, with particular focus on computer-aided usability evaluation. Finally, we describe the ongoing experimental study we undertook with 82 participants, in which a Kansei-based survey with 21 university websites was performed, and outline preliminary results and prospects.

Modeling information space for decision-making in the interaction of higher education system with regional labor market

The paper describes novel approach to constructing decision-making support system based on multi-dimensional model of information space for the process of interaction between higher education system and regional labor market. The particular features of the proposed approach are: accumulation of data with the use of web-scrapping intelligent information system, building multi-dimensional OLAP-warehouse with data consolidation and aggregation, analysis of data based on OLAP technology and deep Data Mining. We provide an overview of up-to-date online data auto-collection technologies, and describe the architecture of the intelligent system accumulating labor market data since 2011. Finally, an example of analysis is presented that allows hypothesizing that labor market is oversaturated by NSTU graduates of certain majors.

HCI Vision for Automated Analysis and Mining of Web User Interfaces

Most techniques for webpage structure and design mining are based on code analysis and are detached from a human user’s perception of the web user interface (WUI). Our paper is dedicated to approaches that instead focus on analysis of webpage’s visual representation – the way it is rendered in different browsers and environments and delivered to the end user. Specifically, we describe the software tool that we built, which takes a WUI screenshot and produces structured and machine-readable representation (JSON) of interface elements as made out by a human user. The implementation is based on OpenCV (image recognition functions), dlib (trained detector for the elements’ classification), and Tesseract (label and content text recognition). To demonstrate feasibility of the approach, we describe application of our analyzer tool to auto-calculate certain measures for a WUI and to predict users’ subjective impressions. Particularly, we assess UI visual complexity, which is known to significantly influence both cognitive and affective aspects of interaction. The results suggest the analyzer’s output is mostly characteristic of the users’ visual perception and can be useful for auto-assessing and comparing WUIs.

Human Capital in the Information Society and the Wage Difference Factors

In the age of e-civilization, human capital that is capable of knowledge-based innovations gets special attention. We in our paper focus on researching new forms of social and labor relations and the effect of education, work experience and certain other factors on the wages. We supplement traditional labor statistics with online data collected and structured with a dedicated software system and with personal character features assessed with psychological diagnostics methods. We highlight the two groups of workers in modern economy -- freelancers and full-time employees -- and outline benefits and disadvantages of remote work, based on quantitative evaluations obtained from the modified Mincer model. Particularly, we attained the values for the effect of education level, work experience, residence location, and personal features on wages in the Siberian Federal Okrug.

Psychometric Factors in Human Capital Research: Identification and Modeling of Employee Groups

Post-industrial economy is shaped by both digitalization, i.e. network-based co-ordination of relations, advanced development of service industry, increase in the number of open innovations, and by deep changes in the role of human and knowledge. The development of Information society can provide competitive advantages for a country in the world economy, as cyberspace enhances intellectual and emotional human resources, broadens cognitive, creative and communication skills, thus allowing boosting the human capital. Our paper is dedicated to studying the factors of education, work experience and personal features (psychometric factors) with respect to their effect on wages and overall identification of employee groups. In this, we employ labor market statistics (both traditional one and collected from online sources with a dedicated software system) and psychological diagnostics methods, which we modified according to our tasks. In our current work, two groups of employees were identified based on the above factors and enhanced Mincer model’s quantitative evaluations: freelancers and full-time employees. We particularly consider the effects of education level and type (including open education), work experience, residence location, and personal features on wages in the Siberian Federal Okrug. Keywords—information society; human capital; freelance; Mincer model; web scraping

Assessing Subjective Quality of Web Interaction with Neural Network as Context of Use Model

Despite certain advances in automation of software quality assurance, testing and debugging remain the most laborious activities in the software development cycle. Evaluation of web interaction quality is still largely performed with traditional human effort-intensive methods, particularly due to the inevitable association of website usability with particular contexts of use, target users, tasks, etc. We believe that testing automation in this field may ultimately lead to better online experience for all and are important in promoting e-society development. We propose to employ artificial neural networks to predict website users’ subjective impressions, whose importance is widely recognized but that are somehow overshadowed by the effectiveness and efficiency dimensions. We justify the structure of the network, with the input layer reflecting context of use, while the output layer consisting of the subjective evaluation scales (Beautiful, Evident, Fun, Trustworthy, and Usable). The experimental session with 82 users and 21 university websites was undertaken to collect the evaluation data for the network training. Finally, we verify the validity of the model by comparing it to a certain baseline, analyze the importance of the input factors, and provide recommendations for future evaluations-collecting sessions.