329 Matching Results

Results open in a new window/tab.

Understanding and Co-Creating  Sustainable Climate Resilience in Society (open access)

Understanding and Co-Creating Sustainable Climate Resilience in Society

How can Knowledge and Information management professionals contribute to co-creating sustainable climate resilience in society? What are the sustainable climate-resilient futures we want to see take place in a global, interdependent world, and what is the Knowledge and Information management professionals’ role in making these scenarios happen? K&I Management professionals have essential roles to play in addressing the crises, risks, and urgency that our environment is facing. Their work as teachers, researchers, and experts in processing data and knowledge are crucial to how our world will develop in the incoming decades. In this interactive workshop called knowledge Huddle – developed by GRASP Network – participants are stimulated in a creative way with the use of art to reflect on the above-mentioned questions. Art helps us to imagine alternative futures, and new scenarios, but not only. It also connects to our feelings and emotions and speeds up creative processes. With this creative approach, the workshop tries to grasp what actions are needed. From individual behavior to organizational, national, and international climate actions, all levels are interwoven and interdependent. In this scene starting from the personal/individual is essential. What can each of us individually do but also how can we collectively make …
Date: June 2022
Creator: Kuusipalo-Maatta, Paula & Maatta, Ossi
Object Type: Text
System: The UNT Digital Library
KM Practices - Motivational Structures  within Tax and Legal Knowledge  Management (open access)

KM Practices - Motivational Structures within Tax and Legal Knowledge Management

This presentation deals with motivational ideas and initiatives for knowledge sharing in the field of tax and legal consulting. Sharing knowledge at all levels and in all formats is often difficult or even impossible due to the need to respect confidentiality and data protection laws. However, the example of PwC Switzerland shows how it is still possible to be a knowledge organization, to create a lively and popular exchange between employees and not to reinvent the wheel every day. We will also shed light on what a well-positioned knowledge management can enable in a crisis situation and how it has moved from behind the scenes to the stage. PwC Switzerland is the leading audit and advisory company in Switzerland. We’re a network of firms in 156 countries with over 295,000 people who are committed to delivering quality in assurance, advisory and tax services. PwC Switzerland has over 3,380 employees and partners in 14 locations in Switzerland and one in the Principality of Liechtenstein.
Date: June 2022
Creator: Otto, Katharina
Object Type: Text
System: The UNT Digital Library
Workgroup KM  Certification (ISO  30401/DIN SPEC  91443) (open access)

Workgroup KM Certification (ISO 30401/DIN SPEC 91443)

New standards (ISO 30401 and DIN SPEC 91443) provide new momentum for knowledge management initiatives. Certification becomes a priority to demonstrate KM-proficiency on an organizational level to external partners (customers, suppliers, future employees) as well as to create a sense of urgency within organizations, in particular among organizational leadership. In this workshop, a joint working group of two independent communities and a research organisation, the Association of Knowledge Management for German speaking countries and the Federal Association of Intellectual Capital Statements e.V. presents the framework of DIN SPEC 91443 which was designed specifically for SMEs to operationalize the KM requirements of ISO 30401. The general objective of the workshop is to reflect on suitability of various instruments of KM, in particular the Intellectual Capital Statement – Made in Germany, to deliver on the standards requirements. Looking at some practical cases, an early-stage concept of auditing structures and procedures will be discussed. As a result of the workshop, the organizers aim to propose an auditing framework to certify knowledge management systems as a follow up.
Date: June 2022
Creator: Alwert, Kay; Bornemann, Manfred; Kohl, Holger; Orth, Ronald & Will, Markus
Object Type: Text
System: The UNT Digital Library
Intellectual Capital Management in an Engineering Company (open access)

Intellectual Capital Management in an Engineering Company

Engineering offices are knowledge companies. Systematically maintaining and developing the existing knowledge in and as a company is therefore an essential task of corporate management. Knowledge management at ZM-I includes a training system (ZM-I Campus), a digital knowledge database (intranet) and workshops. For a systematic analysis and evaluation of the company's intellectual capital, a bi-annual certified intellectual capital report is carried out. For this purpose, employees from all divisions and locations come together to systematically analyse and evaluate the intellectual capital in the company. The intellectual capital statement is thus a fundamental management tool at ZM-I. ZM-I is an independent, medium-sized civil engineering company organised as a partnership for demanding projects. Based in Munich, the group has grown continuously and is now active throughout Germany. As a structural engineering company with a focus on statics, ZM-I offers consulting, planning, testing and assessment services in the field of civil engineering.
Date: June 2022
Creator: Mühlbauer, Christian
Object Type: Text
System: The UNT Digital Library
Knowledge Management Enables the  Transformation from a Global Steel Supplier  to a Technology Service Provider (open access)

Knowledge Management Enables the Transformation from a Global Steel Supplier to a Technology Service Provider

In 2018, voestalpine High Performance Metals decided to significantly invest in Knowledge Management to foster their strategy to expand into new markets. After conducting a thorough global feasibility study involving different stakeholders at multiple locations, a hybrid KM strategy was proposed creating networks of experts and identifying valuable content for re-use supported by a KM Service Portfolio. Members of pilot “communities of Practice” (CoP) were involved in the co-design of KM Services and testing of IT platforms. Once the global IT platform was established, the global roll-out of the CoP’s started. Simultaneously, a global governance structure was designed and implemented as well as a 3-tier measurement approach conceptualized and tested. Today, CoP’s focusing on different topics, are conducting dozens of meetings every month and have identified about a thousand documents to be accessed via their electronic “home base” on Sharepoint. A training program to raise awareness and enable understanding of KM has been initiated on a global scale. The implementation of the KM board brought an important boost in participation and professionalism. To further foster and sustain the world-wide participation in the knowledge-network is the challenge we are facing until knowledge management has become an integral undisputed part of the …
Date: June 2022
Creator: Fratzl, Hubert & Plenkers, Sven
Object Type: Text
System: The UNT Digital Library
Conversation about KM at NASA (open access)

Conversation about KM at NASA

Ed Rogers served as Chief Knowledge Officer with NASA Goddard Space Flight Center from 2003 to 2020. He holds a PhD in HRM from Cornell University and teaches a different universities like University of Alabama, George Washington University and Indian School of Business. Since 2020 he is the principal owner of Mayjer Enterprises (Palmyra, Virginia).
Date: June 2022
Creator: Rogers, Ed
Object Type: Text
System: The UNT Digital Library
Fit to Collaborate in the Otto Group:  Experiences and Perspective (open access)

Fit to Collaborate in the Otto Group: Experiences and Perspective

Juliane Dieckmann, Head of Knowledge Management in the Otto Group Holding, is responsible in the Digital & Consulting division for the group-wide transfer of knowledge and the development of KM products specifically aimed at collaboration in organisational units. She focuses with continuing enthusiasm on the task of making Knowledge Management in the Otto Group fit for the path to digital transformation and putting people at the centre of this. The Otto Group is a globally active trading and services group with around 52,000 employees and a turnover of 15.6 billion euros. With 30 major corporate groups it is present in more than 30 countries. Global group activities and strategic partnerships provide the Otto Group with excellent conditions for know-how transfer and the use of synergy potential. Every day, in the course of its business activities, it deals with the core issues of a modern society, including the future of work, value-oriented action, and the opportunities and challenges of the present.
Date: June 2022
Creator: Dieckmann, Juliane
Object Type: Text
System: The UNT Digital Library
BOSCH Experts Organization:  Experiences & Perspectives (open access)

BOSCH Experts Organization: Experiences & Perspectives

The Bosch Group is a leading global supplier of technology and services. It employs roughly 402,600 associates worldwide (as of December 31, 2021). The company generated sales of 78.7 billion euros in 2021. Its operations are divided into four business sectors: Mobility Solutions, Industrial Technology, Consumer Goods, and Energy and Building Technology. At 128 locations across the globe, Bosch employs some 76,100 associates in research and development, of which more than 38,000 are software engineers. Juergen Ebmeyer joined Bosch in 2005 and is the Corporate Process Owner of the Bosch Experts Organization. He managed change projects to create divisional Centers of Competence and coordinates the Bosch-wide Centers of Competence. Lothar Maier works for Bosch since 2005 and is the IT Infrastructure Application Owner of the Bosch Experts Organization. He additionally supports the Bosch organization with further KM tools and methods as e.g. Expert Debriefing moderation.
Date: June 2022
Creator: Ebmeyer, Juergen & Maier, Lothar
Object Type: Text
System: The UNT Digital Library
Semantic Networks and Knowledge  Management – Context Does the Trick (open access)

Semantic Networks and Knowledge Management – Context Does the Trick

With an increasing flood of data and at the same time changing requirements, making the relevant information available in the right context to a customer and within the company for the respective development-, technical- and service-departments is a key success factor for many companies and at the same time an enormous challenge. The problem with folders and filing data in tree structures is that information about an object is scattered across different aspects. Access to this information is only possible if the respective aspects are known. In a semantic network, each object exists only once, all information about this object comes together at this point and it can still be accessed within different contexts. Such a network can be changed at any time and further aspects can be added as needed: semantic machine learning, in order to be able to react flexibly to new requirements. Semantic networks combine functions of ontologies, topic maps, taxonomies and thesauri. They model complex relationships and directly transform large amounts of structured and unstructured content into networked units of knowledge. In this way, computer-readable and computer-usable knowledge bases are created. The creation and maintenance of semantic networks is demand-driven and interactive between humans and computers. …
Date: June 2022
Creator: Munk, Johannes
Object Type: Text
System: The UNT Digital Library

Semantic Networks and Knowledge Management – Context Does the Trick

With an increasing flood of data and at the same time changing requirements, making the relevant information available in the right context to a customer and within the company for the respective development-, technical- and service-departments is a key success factor for many companies and at the same time an enormous challenge. The problem with folders and filing data in tree structures is that information about an object is scattered across different aspects. Access to this information is only possible if the respective aspects are known. In a semantic network, each object exists only once, all information about this object comes together at this point and it can still be accessed within different contexts. Such a network can be changed at any time and further aspects can be added as needed: semantic machine learning, in order to be able to react flexibly to new requirements. Semantic networks combine functions of ontologies, topic maps, taxonomies and thesauri. They model complex relationships and directly transform large amounts of structured and unstructured content into networked units of knowledge. In this way, computer-readable and computer-usable knowledge bases are created. The creation and maintenance of semantic networks is demand-driven and interactive between humans and computers. …
Date: June 2022
Creator: Munk, Johannes
Object Type: Presentation
System: The UNT Digital Library
Intercultural Perspectives on Information  Literacy - a Report on a Transnational  Project to Foster Information Literacy (open access)

Intercultural Perspectives on Information Literacy - a Report on a Transnational Project to Foster Information Literacy

In an increasingly interconnected world, there is a need to prepare students to become more knowledgeable of different cultures and global matters. Furthermore, in a time which is characterized by disinformation, information literacy becomes more important than ever before. By taking into account transnational perspectives in learning these topics, e-learning provides us with the opportunity to connect students who otherwise would not have the chance to meet and get into knowledge related discourse with each other. In this context, the project „Intercultural perspectives on Information Literacy“ (IPIL) aims to realize a transnational learning community in which students from different countries engage in knowledge building discourse to foster intercultural learning and information literacy. The first IPIL course was conducted in winter term 2019/2020. Since then multiple transnational courses and workshops have been carried out. The last course took place in the winter semester 2021/2022. The didactical structure of IPIL follows a constructivist socio-cultural approach of knowledge building and knowledge creation in which learners from different higher education institutions with diverse cultural backgrounds engage in learning related discourse on topics related to information literacy. To enable such learning, the learning environment is structured into three levels (community, learning cycles, group learning task) …
Date: June 2022
Creator: Griesbaum, Joachim
Object Type: Text
System: The UNT Digital Library

Intercultural Perspectives on Information Literacy - a Report on a Transnational Project to Foster Information Literacy

In an increasingly interconnected world, there is a need to prepare students to become more knowledgeable of different cultures and global matters. Furthermore, in a time which is characterized by disinformation, information literacy becomes more important than ever before. By taking into account transnational perspectives in learning these topics, e-learning provides us with the opportunity to connect students who otherwise would not have the chance to meet and get into knowledge related discourse with each other. In this context, the project „Intercultural perspectives on Information Literacy“ (IPIL) aims to realize a transnational learning community in which students from different countries engage in knowledge building discourse to foster intercultural learning and information literacy. The first IPIL course was conducted in winter term 2019/2020. Since then multiple transnational courses and workshops have been carried out. The last course took place in the winter semester 2021/2022. The didactical structure of IPIL follows a constructivist socio-cultural approach of knowledge building and knowledge creation in which learners from different higher education institutions with diverse cultural backgrounds engage in learning related discourse on topics related to information literacy. To enable such learning, the learning environment is structured into three levels (community, learning cycles, group learning task) …
Date: June 2022
Creator: Griesbaum, Joachim
Object Type: Presentation
System: The UNT Digital Library
Localizing Knowledge Management with Country Knowledge Programming in the Asia and Pacific Region (open access)

Localizing Knowledge Management with Country Knowledge Programming in the Asia and Pacific Region

This paper seeks to identify factors that contribute to improving the way the Asian Development Bank (ADB) understands and responds to the knowledge needs of its developing member countries (DMCs). The COVID-19 pandemic and continuing uncertainties highlighted the importance of knowledge management in helping governments in DMCs make well-informed decisions to respond to the crisis and work for recovery. As ADB’s DMCs are at various stages of development, it is essential to identify, differentiate and manage knowledge that can advance these countries to achieve development objectives and part of that the Sustainable Development Goals (SDGs). ADB’s Strategy 2030 aims to strengthen its country-focused approach using the Country Partnership Strategy (CPS) as the main platform to provide lending and knowledge to its DMCs. An important part of the CPS is the country knowledge plan (CKP) that outlines how ADB is supporting the country’s knowledge needs over the CPS period. This paper discusses the significant enhancements introduced to make the CKP process ‘dynamic’, client-driven while supporting national knowledge agendas. From ADB’s perspective, knowledge is an investment and an asset, and it is important to ensure that its DMCs are getting the most return from this capital.
Date: June 2022
Creator: Jose, Mary Jane Carangal-San; Raman, Vivek & Roth, Susann
Object Type: Text
System: The UNT Digital Library
Research Teams: Fostering Scholarship  and Practice (open access)

Research Teams: Fostering Scholarship and Practice

This workshop is presented by members of a University of North Texas research team. First, the team will overview their experience as members of the research team and share experience in areas such as trust formation, team roles, productivity, work-life balance, faculty-students interaction, peer and faculty mentorship, dissertation preparation, and job seeking. Second, the workshop will discuss and brainstorm how this format can be implemented for organizations both with faculty-student teams and with peer-directed teams. Finally, successes and challenges are openly discussed with audience.
Date: June 2022
Creator: Allen, Jeff M., 1968-; Khader, Malak; Njeri, Millicent & Rosellini, Amy
Object Type: Text
System: The UNT Digital Library
From Digital Twins to Digital Selves and Beyond: Perspectives for Social Modeling in a Trans-Humanist World (open access)

From Digital Twins to Digital Selves and Beyond: Perspectives for Social Modeling in a Trans-Humanist World

We discuss the engineering of so-called ‘humanoid’ socio-technical settings mimicking cognitive and social skills. We discuss the influence of Artificial Intelligence (AI) on digital selves. We introduce relevant approaches to awareness and the emergence of consciousness in artificial agents. We define mandatory properties of digital selves so that digital societies can be modeled and we present our thoughts on the role of KM in a trans-humanist world.
Date: June 2022
Creator: Barachini, Franz & Stary, Christian
Object Type: Text
System: The UNT Digital Library
Organizations as Quantum  a Metaphor to Prepare for Proliferated Quantum Supremacy (open access)

Organizations as Quantum a Metaphor to Prepare for Proliferated Quantum Supremacy

Quantum physics surpasses human imagination. It totally contradicts everyday experiences. Even literal and mathematical explanations cannot substitute for a non-intuitive behavior that puzzles many of us. Commercial quantum computers are in reach within this decade. History provides examples that the appearance of new technologies brought metaphors to life that may explain up to that point poorly understood knowledge domains (e.g., "brains are hardware while thoughts are software"). The author describes why the proliferation of quantum computers will be accompanied by the rise of metaphors that explain quantum effects. For one, those might help to better communicate about the best use and consequences of quantum computers. But those metaphors could also shed light on organizational performance at all. To show a practical example, the author proposes how the use of quantum metaphors could help organizations to prepare for the quantum computing era - without being exact in predictions about most likely technical implementation of quantum capabilities.
Date: June 2022
Creator: Holtel, Stefan
Object Type: Text
System: The UNT Digital Library
Learning to Support Computable  Biomedical Knowledge (CBK) (open access)

Learning to Support Computable Biomedical Knowledge (CBK)

Using technology and e-publication formats, Mobilizing Computable Biomedical Knowledge (MCBK) aims to promote learning health services and reduce time to get healthcare products to patients. This paper summarizes developing a community of practice (CoP) experts, providing educational activities, and plans for sustainable Open Educational Resource (OER) materials online as funded by an IMLS grant.* Activities were part of a pilot training for 20 knowledge managers, students and professionals in Library and Information Science (LIS) in December, 2021. The educational goal is to introduce MCBK concepts globally and and to promote KM and LIS leadership in designing healthcare repositories and accessing information. Beyond text and graphics, CBK publications provide encodable prediction models and computable information in electronic media. It is “dynamic knowledge” in open access formats. Such computable knowledge allows readers to evaluate and validate data or tools that may, for example, compute a risk score for infection. Analysis and review by e-journal readers may reduce time to implementation in healthcare systems or K2P (knowledge to performance) from years with traditional peer- reviewed publications to months using electronic, open access. The summaries and exercises in this paper are from presentations during the pilot training, which were assessed and recommended by student …
Date: June 2022
Creator: Swain, Deborah & Cunningham, Christopher
Object Type: Text
System: The UNT Digital Library
Increasing Information Certainty for Post-Traumatic Growth (open access)

Increasing Information Certainty for Post-Traumatic Growth

Trauma, and its associated effects, can be conceptualized as a period of information uncertainty. The natural psychological response to trauma is a period of post-traumatic stress. Trauma occurs when an existing knowledge base has been challenged. Any event that challenges important components of an individual’s assumptive world is said to be traumatic. This post-traumatic period is akin to many theories and concepts in information science including uncertainty reduction, Everyday Life Information Seeking, Sensemaking Theory, Making Meaning and Anomalous States of Knowledge. One possible outcome after the post- traumatic period is post-traumatic growth. Researchers agree post-traumatic growth primarily occurs across one or more of the following domains: personal strength, new possibilities, relating to others, appreciation of life and spiritual change. That is, people affected by trauma tend to grow when they find new or additional paths of information certainty.
Date: June 2022
Creator: Bank, Nicole & Allen, Jeff M., 1968-
Object Type: Text
System: The UNT Digital Library
Psychiatric Disorders Among Opioid Dependents: Socioeconomic and Gender Difference (open access)

Psychiatric Disorders Among Opioid Dependents: Socioeconomic and Gender Difference

The psychiatric disorder rates among opioid dependents have alarmingly increased over the last decades, and these disorders are higher for women than men and higher for individuals in low socioeconomic groups. Previous knowledge asserted that opioids had no addictive or harmful effects that could lead to psychiatric disorders, but the recent discovery of opioid-related knowledge reversed the existing belief. The purpose of this research is to discover how the new knowledge has changed regarding psychiatric disorders from opioids between men and woman and across socioeconomic groups. In order to uncover these changes, the research data is obtained from the Treatment Episode Data Set (TEDS) in 2007 and 2017. While the percentage of psychiatric disorders among opioid dependents is higher for women than men, unexpectedly the growth rate of psychiatric disorders for men is much faster than women. As such, the socially constructed conventional knowledge that psychiatric disorders are women’s illness will change in the near future. Congruent with existing knowledge, psychiatric disorders in this dataset are higher for underprivileged brackets such as those with low education, unemployed, separated, divorced, and widowed people, and Medicaid recipients, and the growth rates for these groups are steeper than their counterparts.
Date: June 2022
Creator: Kim, Yong-Mi
Object Type: Text
System: The UNT Digital Library
An Ontology and a Reasoning Approach For Evacuation in Flood Disaster Response (open access)

An Ontology and a Reasoning Approach For Evacuation in Flood Disaster Response

Managing flood-related data to assist in the disaster management is a critical process of high importance during a flood disaster. These data are heterogeneous and can be provided from different data sources, and integrating them is a challenging task which allows to infer new information that helps in limiting the consequences of a flood. In this paper, we propose a novel approach that manages heterogeneous flood-related data based on semantic web techniques and helps in limiting the damage caused by floods. We first propose an ontology that is used to formally describe the flood-related data, and we build our knowledge graph through integrating heterogeneous data using the proposed ontology. Then, we propose a reasoning approach using SHACL rules to infer new information that helps in managing the flood disaster or in anticipating future events. The experimental evaluations of our proposed approach are conducted on a real case study in the frame of flood disaster management with the aim of generating evacuation priorities. The results show that it succeeds in managing heterogeneous flood-related data and generating evacuation priorities in a very short time.
Date: June 2022
Creator: Daher, Julie Bu; Huygue, Tom; Stolf, Patricia & Hernandez, Nathalie
Object Type: Text
System: The UNT Digital Library
Metadata Practices of Academic Libraries  in Kuwait, Oman, and Qatar: Current  State, Risks, and Perspectives for  Knowledge Management (open access)

Metadata Practices of Academic Libraries in Kuwait, Oman, and Qatar: Current State, Risks, and Perspectives for Knowledge Management

Developing, implementing, and managing metadata is crucial to successful knowledge management, and academic libraries have traditionally played a central role in these activities. The Arabian Gulf countries are underrepresented in the existing research into library metadata practices. This exploratory study used semi-structured interviews of metadata managers at 8 universities with the goal of developing understanding of the current state of metadata practices, including descriptive cataloging, identity management, and knowledge organization in academic libraries of three Arabian Gulf countries (Kuwait, Oman, and Qatar), as well as potential future developments to facilitate discovery of resources. Findings provide insights into this previously under-researched area and contribute to understanding of knowledge management and risks on a global scale.
Date: June 2022
Creator: Zavalina, Oksana & Aljalahmah, Saleh
Object Type: Text
System: The UNT Digital Library
Using Data Visualization Tools to Mitigate the Influx of Information in Organizations (open access)

Using Data Visualization Tools to Mitigate the Influx of Information in Organizations

Considerable research has been conducted on the topic of information overload using different approaches, from marketing and customer demand to information technologies and sciences, and even among mental health professionals. In business the critical question is how does information overload impact processes, operations, and profitability, and how can data visualization help to solve issues with data management and consumption in organizations. The ability to quickly and effectively process information and make decisions equates to organizational survival in a dynamic, knowledge-based economy where all segments of society are heavily affected by information technologies and systems and data management industries. The growing number of systems apparatuses challenges both individuals and organizations, resulting in reports of fatigue and experiences that compromise successful performance. The objective of this literature review is to discuss how data visualization tools help address information overload and optimize decision making and the business intelligence process in organizations. It concludes that data visualization, indeed, is critical in helping individuals capture, manage, organize, visualize, and present understandable data, but that decision making is affected by cognitive factors that interfere with data processing and interpretation in decision makers.
Date: June 2022
Creator: Merlo, Tereza Raquel
Object Type: Text
System: The UNT Digital Library
Application of Big Data Analytics in Precision Medicine: Lesson for Ethiopia (open access)

Application of Big Data Analytics in Precision Medicine: Lesson for Ethiopia

Precision medicine is an emerging approach for disease treatment and prevention that considers individual variability in genes, environment, and lifestyle for each person. Big data analytics (BDA) using cutting-edge technologies helps to design models that can diagnose, treat and predict diseases. In Ethiopia, healthcare service delivery faces many challenges specifically in relation to prescribing the right medicine to the right patient at the right time. Thus, patients face challenges ranging from staying on treatment plans longer, and then leaving treatment, and finally dying of complications. Therefore, the aim of this paper is to explore the trends, challenges, and opportunities of applying BDA in precision medicine globally and take lessons for Ethiopia through a systematic literature review of 19 peer reviewed articles from five databases. The findings indicated that cancer in general, epilepsy, and systemic diseases altogether are areas currently getting big attention. The challenges are attributed to the nature of health data, failure in collaboration for data sharing, ethical and legal issues, interoperability of systems, poor knowledge skills and culture, and poor infrastructure. Development of modern technologies, experimental technologies and methods, cloud computing, Internet of Things, social networks and Ethiopia’s government initiative to promote private technological firms could be an …
Date: June 2022
Creator: Woldemariam, Misganaw Tadesse & Alemneh, Daniel Gelaw
Object Type: Text
System: The UNT Digital Library
Stock2Vec: An Embedding to Improve Predictive Models for Companies (open access)

Stock2Vec: An Embedding to Improve Predictive Models for Companies

Building predictive models for companies often relies on inference using historical data of companies in the same industry sector. However, companies are similar across a variety of dimensions that should be leveraged in relevant prediction problems. This is particularly true for large, complex organizations which may not be well defined by a single industry and have no clear peers. To enable prediction using company information across a variety of dimensions, we create an embedding of company stocks, Stock2Vec, which can be easily added to any prediction model that applies to companies with associated stock prices. We describe the process of creating this rich vector representation from stock price fluctuations and characterize what the dimensions represent. We then conduct comprehensive experiments to evaluate this embedding in applied machine learning problems in various business contexts. Our experiment results demonstrate that the four features in the Stock2Vec embedding can readily augment existing cross-company models and enhance cross-company predictions.
Date: June 2022
Creator: Yi, Ziruo; Xiao, Ting; Kaz-Onyeakazi, Ijeoma; Ratnam, Cheran; Medeiros, Theophilus; Nelson, Phillip et al.
Object Type: Text
System: The UNT Digital Library