Yueting Chai

A look at task-switching and multi-tasking behaviors

The task-switching behavior study was performed on a big data set with 3000 subjects.Star structure was found in subjects task-switching behaviors on computer usage.Power-law distributions were found in task-switching and multi-tasking behaviors.A few key tasks were found that count for the most of the task-switching behaviors.Younger generations are more accustomed to task-switching/multi-tasking activities. In the past decade, online applications and working platforms, instant messengers and online social networks have become progressively mainstream with the introduction of easily accessible internet and commercially available technological devices. One of the results of the explosion of these online applications and working platforms is the emergence of more and more multitasking activity, that people are doing several types of tasks on computers or mobile devices simultaneously. In this paper, we present an intensive study on a dataset which contains over 15 million computer operation log records from 3000 random selected subjects. The dataset gives us an opportunity to look into people real-world task-switching behavior of computer usage in a very large scale. We explore the characteristics and the star structure of peoples general task-switching and multitasking behaviors on a group level. Our experiments show the existence of Power-law distributions in subjects task-switching activities, which suggests that most of the task-switching events in the dataset are around to a very small number of some hub computer-based tasks. Those top hub tasks include online chatting, browsing internet, document editing and online shopping. At last, the paper explored the interplay between subjects age attribute and their active level during task-switching activities in a quantitative way.

A Secure Model for Electronic Contract Enactment, Monitoring and Management

With the rapid development of e-commerce, electronic contracts are not secure enough for both businessmen and government, due to the absence of authentication, integrality, non-repudiation, uniform specification and centralized management. We propose that e-contract should involve digital signature as its primary element, and the E-Contract Record Center should be introduced into the management of e-contract. In this paper, we present a model for electronic contract enactment, monitoring and management based on the security view. Additional, a framework for e-contract management is presented with the purpose of regularizing the e-commerce market, protecting businessmen’s right and interests, and facilitating governmental monitoring of the e-commerce market, where we define a lifecycle of e-contract and recounted the establishment process of e-contract. Finally, we present implement architecture for the model and framework given above, which give the service and module design of two systems: e-contract system and ECRC system.

A novel E-Invoice Framework towards data-oriented taxation system

With the rapid development of Internet and information technology, our daily lives are changing from time to time. Among these changes, on-line shopping brings a lot of convenience. How to make taxation system satisfy on-line commerce activities is a new challenge. This paper proposes an E-Invoice Framework, which aims at supporting data-oriented taxation systems which focuses more on direct tax-related data (e-invoice data) than taxpayers. The proposed framework includes a Secure E-Invoice Issuing System to generate e-invoices automatically and an E-Invoice Information Management Platform to provide e-invoice related services. In this way, the efficiency of taxation could be great improved and the labor of taxation could be obviously reduced.

Opportunity cost based constraint model for transaction credit evaluation

Evaluating repayment ability and willing has abstracted a lot of attention from various aspects, such as finance, marketing, economics, mathematics and information science. Various credit evaluation methods emerge from time to time. Despite of the differences among these methods, constraints cannot be absent. This paper proposes an opportunity cost based transaction credit evaluation constraint model which tries to organize constraints in a unique way. And the proposed model also demonstrates its feasibility by analyzing a well-known basic transaction credit evaluation model.

A method for online retail sales estimation based on semantic features of web pages

Currently, e-commerce is being applied more and more widely in daily life. However, how to give an accurate, real-time and low-cost estimation of online retail sales is still a difficult problem from both academic and industrial aspects. This paper presents an efficient method for the estimation of online retail sales that is characterized by an order detection algorithm embedded in distributed clients to detect transaction amounts of successful orders. The proposed order detection algorithm is a kind of logistic regression classifier based on web semantic features, which divides web pages into three categories: ordinary pages, order placement pages and order confirmation pages. A further 10-fold validation is conducted and proves the algorithm is quite effective.

Human Computer Interaction Activity Based User Identification

In this paper, an intensive study is performed on the dataset that contains peoples activity records of computer usage, and a multi-feature based user identification algorithm is developed. It is shown that features generated from users program/process start up history and website access records are extremely effective to differentiate people from each other. The proposed algorithm exploits this fact for identifying user by matching users current behavior features with the history feature library. The experimental results show that the proposed multi-feature based algorithm can achieve 91.2% user identification accuracy.

Localized ambient solidity separation algorithm based computer user segmentation

Most of popular clustering methods typically have some strong assumptions of the dataset. For example, the k-means implicitly assumes that all clusters come from spherical Gaussian distributions which have different means but the same covariance. However, when dealing with datasets that have diverse distribution shapes or high dimensionality, these assumptions might not be valid anymore. In order to overcome this weakness, we proposed a new clustering algorithm named localized ambient solidity separation (LASS) algorithm, using a new isolation criterion called centroid distance. Compared with other density based isolation criteria, our proposed centroid distance isolation criterion addresses the problem caused by high dimensionality and varying density. The experiment on a designed two-dimensional benchmark dataset shows that our proposed LASS algorithm not only inherits the advantage of the original dissimilarity increments clustering method to separate naturally isolated clusters but also can identify the clusters which are adjacent, overlapping, and under background noise. Finally, we compared our LASS algorithm with the dissimilarity increments clustering method on a massive computer user dataset with over two million records that contains demographic and behaviors information. The results show that LASS algorithm works extremely well on this computer user dataset and can gain more knowledge from it.

Infra-marginal analysis model for provision mode selection for e-commerce services

E-commerce has grown extraordinarily since the emergence of the internet, and many types of services are employed to accelerate this process. Service quality and productivity are two critical indicators to evaluate the competitiveness of e-commerce companies. Deciding which provision mode of e-commerce services (buy, sell, or self-provide) to adopt is a key operational strategy issue. This paper investigates the conditions and limitations of e-commerce services optimal supply modes, and proposes a cost oriented infra-marginal model where service demand is considered an exogenous variable due to its non-elastic and unprofitable characteristics. By analyzing the main impact factors of this model, this paper infers provision mode selection strategies, which are determined by four factors: transaction cost, service price, service demand, and competitive advantages. Decision trees are derived from these strategies to help e-commerce companies make appropriate decisions. Finally, the proposed models feasibility is verified by two case studies.

A novel transaction financing model towards electronic commerce

In typical commercial financing scenarios, due to strict credit mechanism financial institutions need to inspect series of qualifications of borrowers, such as fixed assets, debt ratio, market scale, management level, earning capacity, and credit mechanism. In consideration of risk control and revenue maximization, only limited number of companies which have large scale, good balance sheet and perfect credit record can finally get loans. However, from aspect of the whole economy, promising companies tend to lose this kind of financing opportunities, especially for SMEs (small and medium enterprises). Fortunately, supply chain financing analyzes financing problem from view of entire supply chain, rather than single enterprise. It provides financial support to entire supply chain via loaning to hub companies based on their surveys. And in view of quick responding to the rapid changing market, timely financing is of great importance to the whole supply chain, which is essential to online trading velocity and scale. In this paper, an online transaction financing model is proposed on the basis of goods mortgage. In this model, transaction financing platform is deemed as a hub position of the whole supply chain, which acquires consolidated credit loans from financial institutions, and provides online financing services for individual companies within given supply chain. And this platform controls risks by goods mortgage and evaluates dynamic credit lines by trade records. In this way, companies could get loans in a timely manner, the platform profits from interest rate differential, and furthermore online trading velocity and scale will be promoted in return. Keywordselectronic commerce; transaction financing; goods mortgage; dynamic credit strategy; interest rate