An Empirical Comparison of Machine Learning Techniques: Focused on Crunchbase

 Glynnis Tan and Yesica Espinoza-Xitumul/ Dr. Byeonghwa Park

Kean University

tangly@kean.edu

Abstract

    

    According to Abraham Lincoln, the best way to predict the future is to create it. This is what Crunchbase strives to do. Crunchbase is a prospecting platform for dealmakers who want to search less and close more in terms of small businesses. With powerful features including personalized account recommendations, lists, territory preferences, intent signals, news notifications, and advanced search, Crunchbase aims to make it easy to target the right opportunities at the right time for small businesses. Since the prediction of the success or failure of a business venture has been something that people have struggled with for centuries, companies like Crunchbase’s best-in-class private company data offers insight into target companies’ teams, funding status, growth trends, tech stacks, web traffic, investments, and more, to personalize the outreach and increase engagement.  In this study, we will provide a comparison of different machine learning techniques for classification in startup’s outcomes (i.e. exit via IPOs, acquisition) from Crunchbase data. More than thousands of startups' data sets, including startup funding and status, can be found on Crunchbase. By using machine learning techniques, specifically logistic regression using Crunchbase data, we hope to predict the key factors needed for the success rate in a startup business. The machine learning techniques will be compared to see which technique is best to use and which one shows better and more accurate results. 

Introduction

        Startups are high-risk companies within their first stages of operations (Bento, 2017). To classify a business as a startup, innovation, the ability to grow, age, and profitability are often considered (Luc, et al. 2020). Over the years, there are many startups who’ve succeeded and obtained a space in the market, while others struggle to have the same results. The issue becomes how can startups know whether their business succeeds or fails? Are there any important factors that determine the success of a startup that they should be aware of? The ability of a startup company can meet its goals or predefined expectations may determine whether the startup is successful (Luc, et al 2020). Logistic Regression is a Machine Learning technique and can be defined as a modeling technique where the dependent variable takes binary values such as “0” and “1” (Bento, 2017). In this case, “0” means unsuccessful, and “1” is successful. Logistic Regression is used to predict the probability of a categorical dependent variable. Based on previous research, consistency with innovation, funding, and financial support are major factors in the success of startups (Okrah, et al., 2018).

Research Question: What variables play a key role in the success or failure of a startup company using the Logistic Regression technique?

Steps

STEP 1: WHAT IS A STARTUP?

    Startups are essentially, a high-risk company within its first stages of operations and usually related to products and services (Bento, 2017). Certain features that categorize a business as a startup include innovation, ability to grow, age, and profitability (Luc, T., Thanh, L., and Phung, N. 2020). 

STEP 2: IDENTIFYING THE FACTORS

    A successful startup depends on whether the business goals or predefined expectations can and are met (Luc, T., Thanh, L., and Phung, N. 2020). Success can be defined as “the measure of performance that occurs when businesses create value for their customers in a sustainable and cost-effective manner” (Luc, T., Thanh, L., and Phung, N. 2020). Survival is taken into account in the criteria of business success and is measured through performance and depends on the capability of the company to continue to operate (Luc, T., Thanh, L. and Phung, N. 2020). 

STEP 3: MACHINE LEARNING

    Competitive advantage takes a part in the success of a startup as it determines and secures them a place in the industry (Skawińska & Zalewski, 2020). Advantages primarily come from intangible factors such as distinctive capabilities and advantages of competence (Skawińska & Zalewski, 2020). 

STEP 4: LOGISTIC REGRESSION

    Research done by Okrah and Nepp, suggests that financial support is relevant for a startup to be innovative which in cause will affect their success, and is the main reason for bankruptcy and the exit of the startup from the market (Okrah & Nepp, 2018).

Methods

1. Defined a startup company and identified the factors that affect startup companies and their success.
2. Used Machine Learning technique, specifically Logistic Regression to predict which factors aided in the success of a startup company.
3. Built a Logistic Regression Model to identify the factors that affected startup companies. 

Results

    By applying the Logistic Regression technique to the Crunchbase dataset, we were able to eliminate any outliers that served no purpose in determining the success of a startup. We were able to create a model that determined the success and failure of a startup. Our model was able to achieve a 79.5% accuracy in predicting the success and failure of a startup company. 

Conclusion

Our study investigates, by implementing the Logistic Regression technique on the Crunchbase dataset, the algorithm that was used to predict the probability of a categorical dependent variable still needs to be explored.

For future research, we would like to see whether the accuracy of our model will increase or decrease if we tune our model. In addition, we would like to apply different machine learning techniques to compare with the logistic regression technique (i.e. Random Forest).   

References

Luc, T. T., Thanh, L. K. H., & Phung, N. T. K. (2020). Studying the Successor Startup Enterprises—A Case Study of Quang Binh 

Province, Vietnam. Open Journal of Business and Management, Volume 8(Number 4), 1426–1438. https://doi.org/10.4236/ojbm.2020.84091

Okrah, J., & Nepp, A. (2018). Factors Affecting Startup Innovation and Growth. Journal of Advanced Management Science, Volume 6(I), 

34–38. https://doi.org/10.18178/joams.6.1.34-38

Okrah, J., Nepp, A., & Agbozo, E. (2018). Exploring the factors of startup success and growth. The Business and Management Review, 

Volume 9(Number 3), 229–237. https://cberuk.com/cdn/conference_proceedings/2019-07-14-09-58-17-AM.pdf

Skawińska, E., & Zalewski, R. (2020). Success Factors of Startups in the EU—A Comparative Study. Sustainability, 12, 1–28. 

https://doi.org/10.3390/su12198200

Tong, Y., & Saladrigues Solé, R. (2019). An introduction to the study on start-up success. Start Up Notes, 1, 51–66. 

https://doi.org/10.21001/SUN.2019.1.04

Read More

The Effect of Emotional Intelligence and Shared Leadership and the Mediation Effect of Team Cohesion and Team Trust: IMOI Model Approach

Glynnis Tan and Justin Antonio/ Dr. Kihwan Kim

Kean University

tangly@kean.edu

 Abstract

       Shared leadership is a social process shared by members aiming toward the completion of shared goals. The extant literature on shared leadership fails to explore its development systematically. This study investigates shared leadership development via the Input-Mediator-Output-Input model. Our hypothesis states emotional intelligence influences shared leadership and team cohesion and trust will mediate both emotional intelligence and shared leadership. 187 senior undergraduates participated in Capstone for 12 weeks. Longitudinal data were collected at three different times with path analysis to test the hypothesis. Data revealed that team members’ emotional intelligence influenced shared leadership behaviors via the development of team cohesion and trust.

Introduction

Shared Leadership is defined as an emergent team process by the distribution of leadership functions among multiple team members. Research Questions: 1.) Will emotional intelligence influence the development of shared leadership over time? 2.) Will team cohesion mediate the relationship between emotional intelligence and shared leadership over time? 3.) Will trust mediate the relationship between emotional intelligence and shared leadership over time?

Hypothesis Development

Emotional Intelligence:

an individual’s ability to understand and control one’s emotion and understand and motivate others' emotion

Trust:

an expectation or belief toward another person having good intentions to achieve better group performance and requires an interpersonal relationship as a prerequisite condition

Team Cohesion:

characterized as the level of attraction of team members to the team and the desire for members to want to stay in the team

Research Methodology and Experiment

Participants

University seniors with majors in several fields within Business (e.g. Finance, Accounting, Marketing, Management, Global Business) at a mid-sized, mid-Atlantic, U.S. University, competed in an online business simulation (“Capsim”) during their required capstone course.  Participants were assigned to each team to manage a start-up manufacturing technology company over an eight-year simulated period that pragmatically took place within one semester.

Individual survey results, aggregated for each team, measured behaviors that affect team cohesion such as team efficacy, satisfaction, and trust.  A total of 186 students on 47 teams were involved in this longitudinal study.  Of the participants, 50 percent were male and 50 percent female.  The average age of the participants was 24.6 years (s.d. = 5.6), and the average self-reported grade point average (GPA) was 2.91 (s.d. = 0.52).  Each team had four or five members, including at least one marketing, finance, management, and global business major student in order to provide skills to run various functions within the company.  Regarding ethnicity, students self-reported as White (46.8%), Hispanic (20.7%), African American (16.7%), Asian (11.1%), and others (4.8%).  Each team consisted of 4-5 students of diversity by ethnic background, age, gender, and academic major.  

Experiment 

Participants were assigned to teams to compete in a business simulation game as a requirement of their course.  Each team had four or five members and managed a small start-up company, each with exactly the same resources at the beginning of the competition.  For twelve rounds, representing 12 weeks, the teams completed decision-making in the area of R&D, Marketing, Production, and Finance. After each round, the simulation game produced various outcomes such as profit, stock price, market share, and debt ratio, which reflect team performances.  The simulation mimics the general processes of running a manufacturing corporation, whose key decisions should bring bigger profits to the company in a competitive environment. Thus, participants strived to make a bigger profit each round, beating its competitors. The teams’ performance was measured by the number of profits and the rankings based on the profit amounts. The result of the simulation accounted for 20% of students’ final grades.  During the experiment, we collected data at four different time points. The participants played the games for 12 rounds. Each round is equivalent to one week. We collected the data of emotional intelligence at the beginning of the game (T0), self–efficacy, and trust after week (T1), team cohesion after week 8 (T2), and the data on team performance after week 12.

Measures

Shared Leadership

We measured the participants’ perceptions of shared leadership using the scale developed by …… (       ). The scale has 10 items. The items were assessed on a five-point Likert scale, with “1” representing “strongly disagree” and “5” representing “strongly agree.”  The individual scores for each team were averaged to derive a final measure of the group level of shared leadership.

The values of ICC (1) and ICC (2) were 0.20 and 0.52 for team cohesion, F = 2.61, p< 0.01. The median rwg(j) value was 0.72. These outcomes justified the aggregation of individual scores of team cohesion to derive the team level score. 

Emotional intelligence.  We measured the participants’ perceptions of their own emotional intelligence using the scale developed by Law et al. (2004, α = 88; the current study α = .81).  The scale has 16 items and reflects four dimensions of emotional intelligence: Self-emotion appraisal, Others’ emotion appraisal, Use of emotion, and Regulation of emotion.  The items were assessed on a five-point Likert scale, with “1” representing “strongly disagree” and “5” representing “strongly agree.”  The individual scores for each team were averaged to derive a final measure of the group level of emotional intelligence. The individual scores for each team were averaged to derive a final measure of trust. The values of ICC (1) and ICC (2) were 0.19 and 0.64 for trust, F = 2.02, p<0.05. The median rwg(j) value was 0.72. These outcomes justified the aggregation of an individual score of team efficacy to derive the team-level score.

Trust.  The perception of trust in team members was measured by using an eleven items scale (McAllister, 1995). The reliability of this scale for the current study was 0.91, showing high reliability of the measure.  The participants responded to these six questions using a five-point Likert scale, with “1” representing “strongly disagree” and “5” representing “strongly agree”.  The individual scores for each team were averaged to derive a final measure of trust.

The individual scores for each team were averaged to derive a final measure of trust. The values of ICC (1) and ICC (2) were 0.19 and 0.64 for trust, F = 2.02, p<0.05. The median rwg(j) value was 0.72. These outcomes justified the aggregation of an individual score of team efficacy to derive the team level score.

Team Cohesion.  The perception of team cohesion was measured by the scale developed by Michalisin and his colleagues (2004, α = 0.84; the current study α = 0.81).  The scale has six items and measures the extent to which team members stick together and remain united in the pursuit of a common goal. The participants responded to these six questions using a five-point Likert scale, with “1” representing “strongly disagree” and “5” representing “strongly agree”.  The individual scores for each team were averaged to derive a final measure of team cohesion. The values of ICC (1) and ICC (2) were 0.20 and 0.52 for team cohesion, F = 2.61, p< 0.01. The median rwg(j) value was 0.72. These outcomes justified the aggregation of individual scores of team cohesion to derive the team-level score.

Team Satisfaction.  The seven items for individual satisfaction with the team were adapted to this study from Peeters et al. (2006).   The reliability score of the scale was 0.85, showing high reliability.  The items were assessed on a five-point scale, with “1” representing “strongly disagree” and “5” representing “strongly agree.”  The team score was aggregated by averaging individual scores for each team. The values of ICC (1) and ICC (2) were 0.12 and 0.69 for team satisfaction, F = 2.98, p< 0.01. The median rwg(j) value was 0.72. These outcomes justified the aggregation of individual scores of team cohesion to derive the team-level score. 

Team Performance.  We measured each team’s performance based on the amount of profit achieved at each round and the team’s cumulative profit at the end of the competition.  The simulation generated numerous financial outcomes including net profit, stock price, market share, turnover rate, debt ratio, etc.  Among these indicators, we selected net profit as a major indicator of team performance because we consider it a major measure that best represented the nature of running a business.  After each round (each week), the teams received documented feedback (e.g., simulation round report) about their performance and a partial class score based on each round’s outcome.  This competitive setting motivated students to actively engage in each round, making the research study closer to a real-world setting.  

Findings

• Emotional Intelligence directly promoted shared leadership. (H1 supported)
• EI indirectly promoted shared leadership via the development of trust and team cohesion. (H2 &3 supported)
• Share Leadership has a positive impact on team satisfaction and team cohesion (H4 supported)

Conclusion

    The current study investigated the development process of shared leadership by employing IMOI model process, revealing that team members’ emotional intelligence has influenced shared leadership behavior via the development of team cohesion and trust. The adoption of IMOI model enables to highlight the dynamic process of the shared leadership development process. Also, IMOI model enables to open new research opportunities by investigating more possible input variables and emergent states, the interactions between emergent states and shared leadership, and the reciprocal relationship between team performance and shared leadership. In this section, we will discuss practical implications, limitations, and future research issues.

References

*Avolio, B. J., Jung, D., Murry, W., & Sivasubramaniam, N. 1996. Building highly developed teams: Focusing on shared leadership process, efficacy, trust, and performance. In D. A. J. D. A. Beyerlein & S. T. Beyerlein (Eds.), Advances in interdisciplinary studies of work teams: 173-209. Greenwich, CT: JAI Press.

Carson, J.B., Tesluk, P.E. & Marrone, J.A. (2007), “Shared Leadership in Teams: An Investigation of Antecedent Conditions and Performance”, Academy of Management Journal, 50(5): 1217-34.

Chia-Yen (Chad) Chiu, Owens, B. P., & Tesluk, P. E. (2016). Initiating and utilizing shared leadership in teams: The role of leader humility, team proactive personality, and team performance capability. Journal of Applied Psychology, 101(12), 1705–1720.

DeRue, D. S. (2011). Adaptive leadership theory: Leading and following as a complex adaptive process. Research in Organizational Behavior, 31, 125–150. http://dx.doi.org/10.1016/j.riob.2011.09.007

Hoch, J. E., Pearce, C. L., & Welzel, L. (2010). Is the most effective team leadership shared? the impact of shared leadership, age diversity, and coordination on team performance. Journal of Personnel Psychology, 9(3), 105-116.

*Hoch, J. E., & Kozlowski, S. W. 2012. Leading virtual teams: Hierarchical leadership, structural supports, and shared team leadership. Journal of Applied Psychology. Advance online publication. Mehra, A., Smith, B. R., Dixon, A. L., & Robertson, B. (2006). Distributed leadership in teams: The network of leadership perceptions and team performance. The Leadership Quarterly, 17(3), 232-245.

Mendez, M. J. 2009. A closer look into collective leadership: Is leadership shared or distributed? Unpublished doctoral dissertation, New Mexico State University, Las Cruces.

Serban, A. & Roberts, A. J. (2016), “Exploring Antecedents and Outcomes of Shared Leadership in a Creative Context: A Mixed- methods Approach”, The Leadership Quarterly, 27(2): 181-99.

Small, E. E., Rentsch, J.R. (2010). Shared leadership in teams: A matter of distribution. Journal of Personnel Psychology, 9 (4), 203-211.

Zhou, W. 2012. Moderating and mediating effects of shared leadership on the relationship between entrepreneurial team diversity and performance. Unpublished doctoral dissertation, City University of New York, NY.

Read More

The “Smart” University: “Smart City” Urban Development Applications

Glynnis Tan/ Dr. Julia Nevarez
Kean University
tangly@kean.edu

Abstract

In this ever-changing world, we live in, technology plays a crucial role. Smart city urban development seeks to bring solutions that center on data to address the many urban challenges that cities face. The purpose of this presentation is to understand how a university model is implemented in smart cities by exploring the potential of Kean University’s involvement in transforming its home city, Union Township, NJ into a smart city. A smart city is generally a technologically advanced urban city that uses communication and informational technologies platforms and infrastructure to increase operational efficiency. The Internet of things (IoT) provides the infrastructure through sensors and applications to collect data that is then analyzed to affect services. The smart city urban development model has been sponsored by university institutions – to a lesser or larger extent – based on urban innovations and services that benefit from university knowledge and resources with governmental agencies and different groups of stakeholders. Examples of smart city approach to urban development in universities can be found in Rutgers University-Newark. Rutgers University developed the Smart and Connected Newark project to enhance Newark residents’ urban lives to foster social inclusion and equity by integrating services offered by the city based on social science research in collaboration with different higher education institutions. The services included are street cleaning, citizen engagement, and bike and scooter share, among others. Sustainability, quality of life issues, and initiatives to deter global warming are some of the most common smart city applications. Building upon integrative literature reviews of Smart Cities, the proposed presentation will examine the role of the university as an anchor institution to facilitate the growth of sustainable practices that impact urban challenges.

  Introduction

The city of Union Township, NJ (Union) is the current home of the state-designated public urban research university- Kean University. The university is positioned as a state-wide leader in research and policy for underserved cities and urban communities. Union may possibly benefit from the university’s efforts and make improvements in areas such as city planning, industries, and integrative use of versatile information, communication, and digital technologies. Essentially, the presence of a quality management, education, research, and societal collaboration with the University could improve the success factors of transforming Union into a “Smart City.” (Anita & Jussila, 2018) Examining an existing University’s “Smart City” project, Rutgers University created the “Smart & Connected Newark” project, a three-year project (10/2020-09/2023) focused on transforming the city of Newark, NJ to incorporate more Smart and Inclusive Service Integration. Within Rutgers University’s Smart & Connected Newark website, the project team summarizes “Smart city services are deeply embedded in modern cities aiming to enhance various aspects of citizens' lives.” (Smart Cities · Smart and Connected Newark Project, n.d.) This writing indicates that implementing smart and inclusive services to the city of Newark will make Newark a Smart City. Following Rutgers University’s project structure may be a quick approach to evaluating Union’s potential in transforming into a “Smart City”, however, there is a lack of conclusive consensus on how to define and classify the new concept of “Smart Cities” as many leaders utilize the term without meeting a particular standard. (Ben Letaifa, 2015) Therefore, this study will utilize existing literature’s framework of Smart Cities as an evaluation factor of Union’s potentiality to become a Smart City. Building upon smart city literature and case study framework by examination of Kean University’s contributions to urban city development, this research may further clarify the importance and role of a University in Smart City development.

 Research Methodology

1. Laying down the framework of Smart Cities

2. University Role and Contribution to Urban Society

3. Comparison of Rutgers’ Smart and Connected Newark Project to Kean University Initiatives

 Research Methodology: Laying down the framework of Smart Cities

This study builds upon an extensive literature review of Smart Cities. Data collected from the literature will serve as a key framework for the indicators of a smart city. The transformation of a city into a smart city commonly centers on implementing technologies, however other economic and social diversity factors are at play that focus on the overall improvement of the city.  

Research Methodology: University Role and Contribution to Urban Society

This study examines how universities are essential. Universities produce smart people who produce smart ideas and in turn produce smart cities. Universities have multiple underlying factors which help provide the highly-skilled labor and technological innovations necessary to drive growth and ensure competitiveness (Drucker & Goldstein, 2007). There is recognition and advocacy for the mutually-beneficial relationships universities and cities can forge around local and regional development (Goddard, 2009; OECD, 2007; Rodin, 2005)

Research Methodology: Comparison of Rutgers’ Smart and Connected Newark Project to Kean University Initiatives

This study compares Rutgers’ Smart and Connected Newark Project and Kean University’s initiatives by showing which areas of their home city need development or what is currently being done within the city.  The study shows how the area in which each city is in, is developing its lifestyles to get a step closer to becoming a Smart City.

Conclusion

Smart Cities are an emerging concept that typically involves progress among people and technology for society. After examining Kean University’s contributions to urban city development to Rutgers' Smart and Connected Newark Project, this research further clarifies the importance and role of a University in Smart City development.  Kean University’s role of a university as an anchor institution to facilitate the growth of sustainable practices that impact urban challenges is evident. In conclusion, the continuation of this development made by Kean University is essential to the transition from an urban city to a Smart city.

Future Work

In the future, Kean University will be implementing other initiatives to improve the lifestyles of the inhabitants in Union, New Jersey. More research on the definition of a smart city, the social science of urban city development, and institutional improvement activities from Kean University is needed to completely categorize Union, New Jersey as a Smart City. However, it is evident that Kean University plays a key role in its development.

 Resources

Anttila, & Jussila, K. (2018). Universities and smart cities: the challenges to high quality. Total Quality Management & Business Excellence, 29(9-10), 1058–1073. https://doi.org/10.1080/14783363.2018.1486552

Ben Letaifa. (2015). How to strategize smart cities: Revealing the SMART model. Journal of Business Research, 68(7), 1414–1419. https://doi.org/10.1016/j.jbusres.2015.01.024

Drucker, J., & Goldstein, H. (2007). Assessing the regional economic development impacts of

universities: A review of current approaches. International Regional Science Review, 30, 20-46.

Falconer, G., &amp; Mitchell, S. (2012, September). Smart City Framework: A Systematic Process for Enabling Smart+Connected Communities. Cisco Internet Business Solutions Group (IBSG).

Goddard, J. (2009). Reinventing the civic university. London: NESTA.

Kean designated as Doctoral University by Carnegie Commission. Kean University. (2022, February 1). Retrieved February 2, 2022, from https://www.kean.edu/news/kean-designated-doctoral-university-carnegie-commission

Koczera, P. (2021, June 15). Smart City and smart campus collaborations move communities forward. Technology Solutions That Drive Education. Retrieved January 14, 2022, from https://edtechmagazine.com/higher/article/2019/08/smart-city-and-smart-campus-collaborations-move-communities-forward

Smart cities. Smart Cities · Smart and Connected Newark Project. (n.d.). Retrieved February 2022, from https://smartcities.rutgers.edu/

NJ Spotlight News. (2022). Kean University launches Center for Economic Development| NJ Business Beat. Retrieved January 18, 2022, from https://www.youtube.com/watch?v=M1z_UQ2FUvo.

Vanolo, A. (2013). Smartmentality: The smart city as a disciplinary strategy. Urban Studies,

51(5), 883–898.

 

Read More