Predicting Employee Occupation Mobility: A Theory-Driven Deep Learning Approach

2023 | Americas Conference on Information Systems | Citations: 0

Authors: li, lun; Sun, Jingyi; Liu, Rong; Lappas, Theodoros

Abstract: Predicting employees' career moves benefits both talents with career planning a ... Expand

Abstract: Predicting employees' career moves benefits both talents with career planning and firms in preparing for the gain and loss of human capital. In this paper, we follow the categorization theory to design an explainable AI artifact for the employee occupational mobility prediction problem. Under a coherent categorization theory framework, three theory-driven components explain different mobility mechanisms. The experimental results approve the effectiveness of this theory-driven approach compared to state-ofthe-art baselines in terms of occupational mobility prediction. Collapse

Topics: IT career software developer explainable artificial intelligence resource allocation LinkedIn

Methods: experiment artificial neural network theoretical contribution BERT

Theories: category theory

Birds of a Feather Lodge Together?: Predicting Review Sentiment Using Social Categorization Theory

2019 | Americas Conference on Information Systems | Citations: 0

Authors: Rhue, Lauren; Daniel, Sherae

Abstract: Reviews drive online sales. In a perfect world, reviews represent a way to make ... Expand

Abstract: Reviews drive online sales. In a perfect world, reviews represent a way to make the market contain less friction. That is, they accurately estimate the value of products without bias from the person writing the review. However, some suggest that factors that do not represent the quality of the product alter the relationship between the review and the sale. We use Tajfel and Turner (1986)'s social categorization theory to develop a model that predicts the sentiment of the review left by a guest. We test our hypotheses using archival data about interactions on AirBnB. We find that when the host and the guest share the same ethnicity with positive stereotypes, the review is more positive. However, if the host and the guest share an ethnicity with negative stereotypes, the review is more negative. The discussion section reflects on these results with respect to theory and offers practical implications. Collapse

Topics: Airbnb database system online review user-generated content electronic market

Methods: archival research sentiment analysis qualitative content analysis statistical hypothesis test literature study

Theories: social identity theory category theory

Towards common reusable semantics

2016 | Information Systems Frontiers | Citations: 0

Authors: Bouabana-Tebibel, Thouraya; Rubin, Stuart H.

Abstract: ... Expand

Abstract: Collapse

Methods: case study machine learning design artifact

Theories: category theory

Foundations of reusable and interoperable facet models using category theory

2016 | Information Systems Frontiers | Citations: 1

Authors: Harris, Daniel R.

Abstract: Faceted browsing has become ubiquitous with modern digital libraries and online ... Expand

Abstract: Faceted browsing has become ubiquitous with modern digital libraries and online search engines, yet the process is still difficult to abstractly model in a manner that supports the development of interoperable and reusable interfaces. We propose category theory as a theoretical foundation for faceted browsing and demonstrate how the interactive process can be mathematically abstracted. Existing efforts in facet modeling are based upon set theory, formal concept analysis, and light-weight ontologies, but in many regards, they are implementations of faceted browsing rather than a specification of the basic, underlying structures and interactions. We will demonstrate that category theory allows us to specify faceted objects and study the relationships and interactions within a faceted browsing system. Resulting implementations can then be constructed through a category-theoretic lens using these models, allowing abstract comparison and communication that naturally support interoperability and reuse. Collapse

Topics: application programming interface extensible markup language knowledge representation electronic health record programming language

Methods: ontological modelling computational algorithm data modeling

Theories: category theory set theory

Towards a Formal Approach to Information Systems Design Theory Using Category Theory

2012 | DESRIST | Citations: 0

Authors: Widmeyer, George R.

Abstract: This paper reports research in progress for the formalization of the notion of i ... Expand

Abstract: This paper reports research in progress for the formalization of the notion of information systems design theory within the framework of formal logic and category theory. The formalization starts with the concept of four relational systems (empirical, subjective, conceptual and formal) that are linked by the four activities for design science research proposed by Venable (2006). Category theory is used as the basis of representing the concept of an information systems design theory as a formal framework by representing each of the four relational systems as either types (theories) or tokens (models). The arrows (morphisms) between the four concepts are explained using Barwise and Seligman’s (1997) definition of “infomorphisms”. The contributions of this research are that it explicates the role of kernel theory (background theory) in information systems design theory and it links this design theory to information fusion and information flow research efforts. It thus provides structures that represent a formalization of the design of information systems. Collapse

Topics: systems design ITIL computer software

Methods: design theory kernel theory design science conceptual modelling computational algorithm

Theories: category theory information theory goal theory flow theory

Conversation Systems for Requirements Engineering

2001 | Americas Conference on Information Systems | Citations: 0

Authors: Hars, Alexander; Zhong, Jiangfan

Abstract: Requirements analysis is one of the most critical problems in information system ... Expand

Abstract: Requirements analysis is one of the most critical problems in information systems development. While many approaches have been proposed, it has proven difficult to bridge the semantic gap between information technology specialists and users who are familiar with the domain. In this paper a new approach to requirements analysis will be presented that uses information technology to communicate with users in natural language. The paper describes the architecture of a prototype system that is based on natural language processing technology. The goal of the system is to engage the user in a productive dialogue about information systems requirements. The system generates questions about systems requirements, parses the users’ answer and then uses the answer to generate additional questions. This approach has two main advantages: dialogues can be adapted to the level of knowledge and the expertise areas of the user. In addition each session results in a knowledge structure that provides a strong foundation for the subsequent steps in requirements specification. Collapse

Topics: requirements elicitation rapid application development

Methods: language model design artifact

Theories: category theory

A Model for Concurrent Heterogeneous Databases

1996 | Americas Conference on Information Systems | Citations: 0

Authors: Montesi, Danilo; Costa, Jose Felix

Abstract: Extended AbstractDuring the seventies centralized databases were dominant. Tech ... Expand

Abstract: Extended AbstractDuring the seventies centralized databases were dominant. Technology and cultural corporate advance changed the focus to decentralized or distributed databases. Distributed databases have a single logical database that is physically distributed and supports just one data model and transaction language [4]. A federated database is a collection of concurrently interacting heterogeneous databases without an attempt to integrate them using a unified schema [7]. This is due to the proliferation of different database systems during the past three decades. It has created difficult problems arising from the need to access concurrent heterogeneous databases through a single data definition and data manipulation language designed under a single data model. There are two important points in the above vision: heterogeneity and concurrency. Collapse

Topics: database system operating system

Methods: database modeling data modeling design science reference modelling

Theories: category theory