LR

docx

School

Minnesota State University, Mankato *

*We aren’t endorsed by this school

Course

613

Subject

Computer Science

Date

Apr 3, 2024

Type

docx

Pages

Uploaded by msameerbd

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points 1. Attend at least two online workshops organized by CESR. See the information in the CESR Workshops document. Discuss five takeaways from the sessions you attended. Provide the following information: a. Title of the session(s) you attended. b. Discuss each of your takeaways. c. How do you plan to use the takeaways in your research in the future? (40 oints) A. “Developing Research Ideas” Key takeaways and plan for future research The workshop deliberated upon the following subjects, which has significant potential to assist in the application of my forthcoming research or studies in a multitude of ways, encompassing both theoretical and practical dimensions. SWOT Analysis: A strengths, weaknesses, opportunities, and threats (SWOT) analysis plays a pivotal role in developing research ideas. The analysis directs researchers to identify their distinctive strengths and skills, giving them a competitive edge. Whether it is expertise in specific areas like renewable energy or healthcare, the research direction is shaped through this analysis. The assessment of available resources and potential limitations informs realistic research objectives. SWOT analysis catalyzes reflection on external factors like shifting regulations or unexpected competition. In planning future research, SWOT analysis ensures a balance of strengths and weaknesses, maximizes opportunities and considers potential threats, ultimately strengthening the research overall. Survey Construction and Validation Survey validation is essential for ensuring the reliability and validity of collected data. The reliability analysis looks at how stable and consistent the measurements are. It considers random errors in the observed score, the sum of the actual score, and the error. The most common type is internal consistency, often measured using Cronbach's alpha. Conversely, content validity assesses the extent to which survey items represent the material that should be covered. This involves expert judgment and a proper developmental process with clearly defined constructs. Factors to consider include ensuring all significant aspects of the construct are covered, the proper proportionality, and the relevance of all survey items to the construct. Finally, construct

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points validity depends on the survey's purpose, with various ways to establish it beyond the scope of this discussion. Qualitative Research: Qualitative research delves into the scientific investigation of non-numeric data, seeking nuanced explanations behind phenomena. Commonly applied at the intersection of people and context, it aims to understand subjects' perspectives and the meanings they attribute to events. Qualitative Data Analysis (QDA) emphasizes seeking meaning and understanding the "big picture" and relies on the researcher as the primary instrument for data collection and analysis, employing inductive reasoning and hypothesis generation. It finds applications across diverse fields such as sociology, psychology, education, communications, anthropology, and ethnography, addressing various purposes like cultural investigations, exploratory analysis, and social validity. Cultural, social, and theoretical factors impact the researcher's perspective, which affects qualitative research orientation. Valid formats include case studies, with coding crucial to managing and analyzing qualitative data effectively. Coding involves highlighting parts of the data to denote salient aspects, allowing for easier access, comparison, and analysis. Various computer software, such as MAXQDA and NVIVO, aids in coding and analysis. Analytic memos document stray thoughts occurring during the research, potentially leading to new avenues of investigation or theoretical models. Quantitative Research Quantitative research involves a systematic approach to examining a phenomenon or event, collecting numerical data, and employing statistical and mathematical methods for analysis. It tests predetermined hypotheses, utilizing surveys, online polls, and questionnaires for data collection. Commonly applied in biology, business, chemistry, economics, marketing, sociology, psychology, and more, quantitative research presents several vital characteristics. It utilizes methodological tools for gathering numerical data, involves extensive sample sizes, employs closed-ended questions, builds upon previous research, and allows data presentation through tables, charts, etc. Advantages and Disadvantages of Quantitative Research

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Quantitative research is a systematic, numerical approach using statistical methods for hypothesis testing, often employing surveys and questionnaires. Widely applied in fields like biology, business, and psychology, it features characteristics such as numerical data gathering, large sample sizes, closed-ended questions, and data presentation through tables and charts. While strengths include replicability and precise methods, they may be criticized for superficiality and biased structures. Standard analyses involve T-tests, ANOVA, and correlations using tools like SPSS, R Studios, Python, and Excel. Research Questions, GAP, and Evaluation Plan Research questions serve as the foundational starting point for quantitative research, offering a clear direction for knowledge development, methodology, and contributions to the field. They are crucial in shaping the research path with precision and focus. Inspiration for generating these questions can be drawn from societal trends, cautious use of anecdotal experience, existing published literature, and empirical material produced in the research process. The process of formulating research questions involves identifying a domain or subject area, narrowing it down to a specific topic, and addressing the gap in the research, considering both theoretical and practical rationales. In empirical research, gap spotting means finding topics that are unclear or not being studied enough and new uses that can add to and expand on existing literature. Developing a robust research evaluation plan is equally critical to ensuring a sustainable research program by outlining how outcomes will be assessed for grants, publications, and overall project success. B. “Qualitative Data Analysis” Key takeaways and plan for future research The workshop deliberated upon the following subjects, which has significant potential to assist in the application of my forthcoming research or studies in many ways, encompassing both theoretical and practical dimensions. What is Qualitative Research: Qualitative analysis encompasses diverse approaches and methods to study the intricacies of natural social life. In contrast to quantitative analysis, which focuses on numerical data and statistical methods, qualitative analysis delves into the richness of non-numeric data. It operates on inductive reasoning, aiming to generate hypotheses rather than test predefined ones, allowing for a more exploratory and nuanced understanding of phenomena. This methodological flexibility is precious for future research endeavors. Qualitative analysis provides a holistic view,

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points capturing the depth and context of social phenomena. Its emphasis on inductive reasoning encourages researchers to explore new perspectives, fostering creativity and generating hypotheses that may shape the direction of future quantitative studies. The complementary nature of qualitative analysis can enhance the overall robustness of research, offering a more comprehensive understanding of complex social realities. The Goal of Qualitative Research: The goals of qualitative analysis are multifaceted, encompassing a range of objectives that contribute to a deeper understanding of social phenomena. First and foremost, qualitative analysis aims at theory development, striving to unearth new insights and perspectives that can enrich existing theoretical frameworks or lead to the formulation of novel theories. Additionally, it is a powerful tool for documenting cultural observations preserving the nuances and intricacies of social practices and behaviors. The method also seeks to provide new insights into the complexities of social dynamics, offering a more holistic understanding of the factors influencing human behavior. Qualitative analysis is also very useful for evaluating programs and policies because it lets researchers figure out how well and what effects different initiatives have in a specific social setting. By addressing these goals, qualitative analysis contributes significantly to advancing knowledge and improving programs and policies, fostering a more nuanced comprehension of the intricacies of social life. Types of Qualitative Research: Qualitative research encompasses various approaches, each tailored to unravel distinct aspects of human experiences and behaviors. 1. Phenomenological Approach: o Focuses on comprehending similar events as experienced by different individuals. o Examples include exploring the concept of "motherhood," the experience of grocery shopping, and the nuances of getting married. It delves into the dull and exciting facets of these shared experiences. 1. Ethnographic Approach: o Rooted in anthropological research, it analyzes the shared experiences of people within the same culture. o Shifts the lens from individual experiences to understanding the broader cultural context. Examples involve studying immigrant communities, hospitals in low- income areas, and highly successful corporations. 1. Case Study: o It aims to comprehend a single event that a specific case, an individual, group, or organization experienced.

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points o Examines unique instances such as the Phineas Gage accident or the effects of new legislation on communities. 1. Narrative Inquiry: o It focuses on transforming observational data into narratives, creating a chronological timeline of experiences. o Examples range from exploring a family's lineage and a student's university career to documenting a country's war history. It seeks to convey stories embedded in data, providing a rich narrative texture to qualitative research. Methods of Qualitative Research Qualitative research employs various methods to capture the intricate nuances of human experiences and societal phenomena. In contrast to quantitative methods, which involve experiments, numeric observations, and surveys with closed-ended questions, qualitative methods delve deeper into the qualitative aspects of research. 1. Interviews with open-ended questions: o It involves a series of questions the researcher administers, allowing participants to provide open-ended responses. o They are widely used in phenomenological approaches, ethnographic studies, observations, and narrative inquiries. o Drawbacks include time consumption and potential language barriers. 1. Observations Described in Words: o It encompasses watching behaviors, social interactions, patterns, and the general environment. o They are applied in ethnographic approaches, case studies, and phenomenological studies. o Challenges include the potential for the researcher to affect participants' behavior, its time-consuming nature, a lack of subject interaction, and susceptibility to researcher biases. 1. Literature Reviews: Exploring Concepts and Theories: o It involves accessing existing data through literature reviews. o It is less time-consuming and commonly used in narrative inquiry. o Limitations include potential researcher bias and the absence of direct subject interaction. Each qualitative method offers a distinct lens through which researchers can explore and understand the depth and context of human experiences, providing valuable insights beyond numerical data. How to Analysis Data and Tools:

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Data analysis in qualitative research involves systematic processes, such as coding, to make sense of complex information. Coding is a crucial method used to organize data, aiming to break it down into smaller, manageable parts and facilitate the comparison of themes. In qualitative analysis, emerging themes, such as mental fatigue, physical stress, hopelessness, and low student engagement, are identified. It is essential to code themes based on context rather than settings to ensure accurate interpretation. Helpful tools for qualitative analysis include: 1. Text IQ in Qualtrics: o A language processing function that organizes topics by comments to provide users with a general starting point for analysis o It is not a substitute for coding but aids in the initial stages of understanding data. 1. Excel: o Provides organizational freedom when coding, allowing researchers to structure and manage their data effectively. 1. MAXQDA: o A Specialized software designed for qualitative and mixed-method analyses, offering advanced features for in-depth data exploration and interpretation. These tools improve the speed and accuracy of qualitative data analysis. They help researchers understand the research context more fully and develop valuable insights. Future research help The workshop focused on the cultivation of research ideas and the application of qualitative research methods. It provided participants with unique insights and practical tactics that have the potential to enhance the efficacy and comprehensiveness of their future research initiatives. The instruction on SWOT analysis offers researchers a systematic framework for identifying and capitalizing on their distinctive strengths, addressing areas of weakness, exploring new opportunities, and effectively managing potential threats. This comprehensive examination not only influences the trajectory of research but also guarantees a pragmatic and knowledgeable establishment of research goals. Additionally, the training emphasizes the crucial elements of survey creation and validation, highlighting the significance of employing dependable and valid techniques for collecting data. Researchers must understand qualitative research to fully understand and appreciate the complexities of human experiences and social phenomena. This includes understanding its goals, classifications, methods, and tools. Include inductive reasoning, hypothesis development, and using different qualitative approaches in the researcher's toolbox. This gives them a broader range of options for dealing with complex research questions. Providing instruction about data analysis methodologies and tools, such as coding and specialist software like MAXQDA, guarantees that researchers can adequately explore and analyze qualitative data. To sum up, this program gives researchers a complete framework and the skills they need to develop essential research ideas, use qualitative methods effectively, and improve their future research projects' overall quality and impact.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points …………………………………………………………………………………………………… 2. State the title of your research project within 8-10 words. Use the best practices for creating effective titles discussed in the class. (10 points) Best practices for creating effective title The process of generating a study title necessitates the careful evaluation of various crucial factors. The title should be succinct and unambiguous, offering a transparent representation of the research's focal point without superfluous intricacy. The inclusion of pertinent keywords plays a vital role in optimizing search engine rankings and enhancing database discoverability. The act of avoiding ambiguity is crucial in order to facilitate a rapid comprehension of the primary message by readers. The pursuit of distinctiveness contributes to the prominence of one's work, hence augmenting its capacity for recognition. The alignment of tone with one's research approach is a significant factor in shaping the overall impression. The incorporation of essential variables and research design particulars provides a comprehensive background. At the same time, the utilization of an engaging title has the potential to attract readers' attention. Striking a balance between conciseness and providing sufficient information, it is generally recommended to maintain an optimal length of approximately ~ 10 words. Incorporating the research question or hypothesis into the title of a study makes a coherent link with the central inquiry of the research. The title creation process necessitates the regular examination and revision of the title, soliciting comments, and assuring ethical depiction [15,17]. In summary:  Keep the title short (~ 10 words)  Keep the title interesting but not misleading  Don’t “trick” people into reading your paper with a misleading title  Use appropriate words (active verbs) from Bloom’s taxonomy  “Words that do not carry information, such as “The...,” “A...,” “On...,” “Investigation of...,” “Study of...,” and “Using” should be omitted from titles.”  Don’t use “people’s names”—unless it is standard nomenclature such as “Fourier transform”  Avoid equations in the title  Don’t use unfamiliar acronyms  The first word and all words except prepositions and articles are capitalized. Predictive Analytics in Accounting Fraud Detection ………………………………………………………………………………………………

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points 3. Identify at least 10 scholarly articles related to your research topic. Use the library databases and Google Scholar to find the articles. Provide the citations in the ACM or IEEE format. Explain your rationale for selecting the articles. How did you discern whether an article you chose had high-impact and was relevant for your research topic? (25 points) Identification of ten scholarly articles on “Predictive Analytics in Accounting Fraud Detection” 1. Data-driven auditing: A predictive modeling approach to fraud detection and classification [16] 2. Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction [11] 3. The integration of forensic accounting and big data technology frameworks for internal fraud mitigation in the banking industry [2] 4. Machine Learning-based Decision-Making Systems, Cloud Computing and Blockchain Technologies, and Big Data Analytics Algorithms in Accounting and Auditing [5] 5. Fraud Detection in Healthcare Insurance Claims Using Machine Learning [9] 6. Detection of fraudulent transactions using artificial neural networks and decision tree methods [6] 7. Fraudulent financial reporting and data analytics: an explanatory study from Ireland [1] 8. Fighting Accounting Fraud Through Forensic Data Analytics [7] 9. Corporate governance fraud detection from annual reports using big data analytics [14] 10. The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature [10] Rationale and relevance for selection the articles: In selecting articles for the research topic "Predictive Analytics in Accounting Fraud Detection," a meticulous evaluation process was employed to ensure high impact and relevance. Firstly, emphasis was placed on articles published in reputable, peer-reviewed journals with a focus on finance, accounting, data analytics, and related fields. Journals with high impact factors and rigorous editorial standards were prioritized. Additionally, the credibility of the authors and their affiliations played a crucial role. The authors of articles with a track record in fraud detection, predictive analytics, or related fields were experts, researchers, or professionals. One more thing that was used to judge the usefulness of each article was how well it fit with the research goals and the main topic of using predictive analytics to find accounting fraud. Articles that provided novel insights, employed advanced methodologies, and addressed contemporary challenges in

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points the field were deemed highly relevant. This stringent selection process aimed to ensure that the chosen articles not only contributed significantly to the existing literature but also provided valuable and impactful insights for advancing research on predictive analytics in accounting fraud detection. The critical considerations for the selection of the articles are:  Relevance and Timeliness:  Transformation of Fraud Detection  Diverse Methodologies:  Practical Implications for Organizations:  Contribution to Literature:  Alignment with Research Objectives:  Diverse Perspectives: …………………………………………………………………………………………………. 4. Using the best practices stated in the article, “Writing a literature review,” write a systematic literature review that provides a cohesive and analytical summary for five of the articles (about 250-300 words per article; up to 1500 words total). Use the “cite while you write” feature of the citation management software to appropriately cite the articles in the ACM or IEEE format. (50 points) Predictive Analytics in Accounting Fraud Detection Introduction In the dynamic and ever-evolving financial landscape, organizations increasingly use sophisticated analytical tools to combat the pervasive threat of accounting fraud. Predictive analytics, a powerful and transformative approach, has emerged as a game-changer in fraud detection. It empowers businesses to proactively identify and prevent fraudulent activities before they inflict significant financial damage [4,7,8]. In recent decades, there has been a growing focus among researchers and practitioners on the issue of accounting fraud as it has become more prevalent and varied [7]. Accounting fraud, a form of financial deception, involves the manipulation of accounting records or financial statements to misrepresent an entity's financial performance or position. This illicit practice can manifest in various forms, including asset misappropriation, fraudulent financial reporting, and corruption, each potentially eroding investor confidence, damaging brand reputation, and inflicting substantial financial losses [8,13]. On the other hand, predictive analytics, a branch of

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points advanced analytics, harnesses the power of historical and current data, employing statistical modeling, data mining, and machine learning algorithms to uncover patterns, predict future trends, and identify potential anomalies. In the context of accounting fraud detection, predictive analytics delves into vast troves of financial data, scrutinizing transactions, account activity, and behavioral patterns to flag suspicious activities that may signal fraudulent intent [7,11,14]. The significance of predictive analytics in accounting fraud detection lies in its ability to transform the traditional reactive approach to fraud prevention into a proactive, data-driven strategy [4,13]. By analyzing vast amounts of data from disparate sources, predictive analytics algorithms can identify subtly anomalies and patterns that may not be readily apparent to human auditors, enabling organizations to uncover potential fraud schemes early on. The predictive analytics approach in fraud detection yields several compelling advantages for organizations. Firstly, it enables early detection of potential fraud schemes, offering the opportunity for timely intervention and the mitigation of financial losses [14]. This proactive stance is particularly advantageous compared to traditional methods that might identify fraudulent activities at later stages. Secondly, continuously refined through data updates, predictive models exhibit improved accuracy over time. This refinement reduces the likelihood of false positives, ensuring that flagged activities are more likely to be actual instances of fraud [11]. Thirdly, proactive fraud detection minimizes the costs associated with investigations, legal proceedings, and reputational damage by identifying and addressing potential fraud early on. This reduction in financial and non-financial costs contributes significantly to overall organizational resilience [7]. Finally, predictive analytics enhances risk management by providing valuable insights into fraud risks. This allows organizations to develop and implement effective risk mitigation strategies, bolstering their overall resilience in the face of potential fraudulent activities. In essence, the proactive nature of predictive analytics transforms fraud detection from a reactive process into a strategic advantage for organizations [8]. The subsequent sections of this paper are organized in the following manner: This part thoroughly examines the extant literature about the identification of predictive analytics and accounting fraud. This review aims to outline commonly used research methods, present findings, discuss their practical implications, suggest new areas for future research, and draw attention to the limitations identified in previous studies. Drawing upon the findings of the literature study, the subsequent parts will generate research inquiries and construct a succinct problem or goal statement. The problem statement will examine the probable consequences of the selected topic and research questions, evaluating the importance of exploring the chosen field. Additionally, this study intends to solve the existing research gaps. The final section of this article will methodically outline the research objectives derived from the clearly stated problem or purpose statement. This will serve as a guide for the subsequent inquiry. Literature Review

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Data-driven auditing: A predictive modeling approach to fraud detection and classification [16] The paper by Nitin Singh et al., "Data-driven auditing: A predictive modeling approach to fraud detection and classification" (2019), explores a pioneering predictive model for fraud detection in auditing, emphasizing its potential to reduce manual intervention, processing time, and costs. The study focuses on the implications of data-driven audit on the moderating role of auditors in principal-agent relationships, providing practical insights into financial reporting, auditing operations, and fraud detection. The predictive model, rooted in classification techniques, identifies anomalous transactions in real-time, offering significant efficiency gains for auditors. The paper contributes to the literature by discussing the operational aspects of financial reporting and auditing, modeling fraud-detection classification, and evaluating the benefits, barriers, and enablers of implementing data-driven audit in companies. Practical implications of the study highlight the model's ability to enhance audit efficiency, accuracy, and independence. By reducing monitoring and contracting costs and disincentivizing fraud, the data-driven approach promotes objectivity, aligning with the evolving landscape of audit practices. The paper reinforces the relevance of agency theory in understanding auditor relationships. It acknowledges the growing use of information technology in analytics and highlights the need for more discussions on audit efficiency and data-driven audit processes. Methodologically, the paper proposes a predictive model, employing classification techniques like logistic regression, neural networks, support vector machines, and decision trees. The results demonstrate the model's effectiveness in reducing manual labor for internal and external auditors, providing real-time fraud predictions and improving audit objectivity. In conclusion, the paper affirms that a data- driven auditing approach, utilizing predictive modeling and classification, significantly improves fraud detection, audit efficiency, and independence. The recommendations for future work underscore the importance of comparative analytics, accuracy tests, dynamic rule development, and the exploration of unsupervised outlier detection methods to further enhance fraud detection capabilities. Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction [11] The article, "Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction" (2017), significantly contributes to the literature on fraud detection by addressing challenges associated with the rarity of fraud observations and the abundance of explanatory variables. The

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points authors introduce and evaluate three data analytics preprocessing methods designed to enhance fraud prediction performance, particularly in the context of financial statement fraud. The study underscores the importance of recognizing the diversity of fraud types, advocating for the subdivision of the general fraud problem to improve prediction accuracy. Drawing from emerging data analytics literature, the paper systematically evaluates the proposed preprocessing methods, demonstrating their effectiveness in addressing data rarity issues. The research not only improves fraud prediction models but also emphasizes the potential benefits for regulatory bodies, such as the Securities and Exchange Commission (SEC), in detecting fraudulent filings and for audit firms in making informed client portfolio decisions. The literature survey provided by the paper highlights its contribution to the existing body of knowledge on fraud detection. It emphasizes the utility of data analytics techniques in improving fraud prediction models and addresses the challenges posed by missing values in variables. The acknowledgment of the potential negative effects of missing data on fraud prediction underscores the need for further exploration in handling such data effectively. Methodologically, the paper evaluates three data analytics preprocessing methods, namely the Observation Under-Sampling (OU) method, Multi- Subset Variable Under-Sampling (VU ), and the Variable Under-Sampling Partitioned ( PVU) method, and their combination, showcasing their impact on fraud prediction performance. The results demonstrate a substantial improvement of approximately 10 percent compared to current techniques. The study suggests that future research should focus on refining data analytics methods for fraud prediction, exploring different PVU implementations, and addressing issues related to noisy and missing data. In conclusion, the paper offers practical insights into improving fraud prediction models through data analytics preprocessing methods. The demonstrated performance benefits underscore the potential for similar approaches in addressing challenges in various settings beyond financial statement fraud, including bankruptcy, financial statement restatements, internal control weaknesses, and audit qualifications. Overall, the research makes a significant contribution to advancing the field of fraud detection and data analytics applications in auditing. Fraudulent financial reporting and data analytics: an explanatory study from Ireland [1] The article by Ahmed Aboud and Barry Robinson (2022) investigates the application of data analytics, machine learning, and data mining in fraud prevention and detection within Irish businesses. The study underscores the underutilization of these techniques and identifies barriers hindering their implementation. It contributes empirical insights to the literature on big data analytics and auditing, emphasizing the need for advanced analytical methods in detecting accounting fraud accurately and early, surpassing the limitations of standard auditing procedures.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Practical implications highlight the necessity for increased adoption of data analytics in Irish businesses to enhance fraud detection, addressing barriers like cost, lack of training, and expertise. The research, based on a survey of 73 Irish businesses, reveals that 87.67% use data analytics for fraud detection, with larger companies exhibiting higher utilization rates. Financial ratio analysis, data mining, and text mining emerge as primary fraud detection techniques, but the perceived effectiveness of specific tools varies among respondents. Despite widespread use of data analytics in general, only 35% of surveyed businesses use it specifically for fraud detection. The study calls for future research to explore barriers in greater detail, assess the effectiveness of different analytic techniques, understand factors contributing to underutilization, and evaluate the benefits in various sectors, suggesting a comprehensive agenda for advancing the field of fraud detection through data analytics in Ireland. Fighting Accounting Fraud Through Forensic Data Analytics [7] Maria Jofre and Richard H. Gerlach's (2018) paper, "Fighting Accounting Fraud Through Forensic Data Analytics," aims to enhance accounting fraud detection by employing machine learning methods to differentiate between fraudulent and non-fraudulent companies while evaluating industry-specific risk indicators. The proposed methodology, utilizing discriminant analysis, logistic regression, AdaBoost, decision trees, boosted trees, and random forests, contributes to the forensic data analytics field. The study underscores the importance of non- static regulatory interventions and global concerns about accounting fraud's threat to financial system stability and market confidence. Practical implications include the assistance provided to public auditors and regulatory agencies in auditing processes, offering more targeted examinations of accounting reports. Machine learning and industry-specific indicators improve fraud detection, benefiting investors and financial analysts. The paper suggests the significance of data-driven approaches and forensic data analytics in identifying signs of accounting fraud, contributing to the development of robust fraud detection systems. The study demonstrates varying model performance across industries, showcasing potential in detecting falsified financial statements. Concluding with future directions, the authors suggest exploring classification threshold selection, addressing class imbalance challenges with alternative methodologies, and replicating the proposed approach in specific economic domains for valuable insights. In essence, the paper adds valuable insights to the literature on fighting accounting fraud through the innovative application of forensic data analytics and machine learning methods. Corporate governance fraud detection from annual reports using big data analytics [14] . Corporate governance and fraud detection are paramount concerns in contemporary business landscapes. This review centers on the groundbreaking work by G. Sudha Sadasivam et al., "Corporate governance fraud detection from annual reports using big data analytics" (2016). The

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points study introduces an automated analysis of annual reports employing the MapReduce paradigm, achieving a remarkable 90% accuracy in identifying fraudulent companies using only 10% to 25% of principal features. The authors address the inefficiencies of traditional audit procedures in detecting management frauds, advocating for a shift toward big data analytics. The paper underscores the significance of annual reports as vital repositories of economic and technical indicators, emphasizing their susceptibility to manipulation. The proposed method leverages Hadoop for data analytics and HBase for storage, improving time efficiency by 85%. Methodologically, the study integrates principal component analysis (PCA) for feature extraction, employing support vector machines (SVM) for classification. Notably, the MapReduce paradigm significantly enhances time efficiency, reducing feature extraction time tenfold compared to conventional methods. Results showcase a robust fraud detection accuracy of 90%, and the system's performance is tested on prominent companies like Infosys and Zenith Birla. The application of the MapReduce paradigm not only improves accuracy but also reduces feature extraction time by 85%. However, the paper lacks specific details on other performance measures. …………………………………………………………………………………………………….. 5. Based on your literature review, identify gaps in previous research that your research may aim to address. Identify at least three research questions from your literature review. Write a succinct problem/purpose statement using your research questions (200-300 words). The problem statement should address the following: 1. What would be the impact of your topic and research questions? (Is your topic worth researching?) 2. What gaps will your research address? (25 points) Research GAP: The literature study has identified some gaps that could be addressed in future investigations. Article Research GAP Corporate governance fraud detection from annual reports using big data analytics. 1. The proposed approach focuses on analyzing annual reports to detect fraudulent companies. However, it does not consider other data sources, such as social media, news articles, or external audits, which could provide additional insights into fraudulent activities. 2. The paper mentions big data analytics and the MapReduce paradigm for efficient analysis. However, it does not provide detailed information on the

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points techniques or algorithms used for feature extraction and classification. 3. The experimental results show a high accuracy of 90%. However, there needs to be more information on the size or diversity of the dataset used for evaluation, which could impact the generalizability of the findings. 4. The paper needs to discuss the problems or restrictions that arise when using Hadoop and HBase to store and process the extracted features. These could be issues with scalability, data consistency, or security. 5. The literature survey in the paper is limited and does not cover a comprehensive range of existing research on fraud detection in annual reports. Data-driven auditing: A predictive modeling approach to fraud detection and classification 1. The study acknowledges a limitation concerning the positive correlation between data size and prediction accuracy, suggesting that further improvements in accuracy could be achieved by expanding the dataset or incorporating additional attributes. 2. The research is constrained by its reliance on data generated from a singular company, potentially restricting the generalizability of the findings. To enhance external validity, future studies could test the model using datasets from diverse companies. 3. Despite efforts to identify and rectify erroneous entries in the dataset, the paper concedes the possibility of minor errors and input-related issues associated with the data. 4. The study recognizes potential types of fraud that the model may be unable to detect. However, it contends that the model's results align with those of auditors, reducing the likelihood that auditors would have identified these fraudulent transactions either. 5. The paper explicitly acknowledges that there are more comprehensive theories to explain the auditing process than a data-driven approach. It recognizes the potential relevance of other economic theories in understanding and enhancing the auditing framework. 6. Note: The limitations outlined in the paper pertain to the chosen research methodology and dataset. They do not necessarily imply deficiencies in the overall approach or the validity of the findings.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction 1. The paper lacks specific details or targeted recommendations for future research, offering only broad suggestions for improvement in fraud prediction through data analytics methods. 2. The potential limitations or challenges associated with the three data analytics preprocessing methods assessed in the study should be thoroughly discussed and addressed in the paper. 3. The paper does not provide a comprehensive analysis of the impact of missing values on the utility of variables in fraud prediction models. 4. The paper does not discuss or critically evaluate the possible problems or limits of partitioning methods in predicting fraud, such as the fact that they might introduce bias by specific subsets. Fraudulent financial reporting and data analytics: an explanatory study from Ireland 1. The study's reliance on a relatively small sample size of 73 Irish businesses may limit the generalizability of the findings. 2. The closed-question format employed in the survey might have limited the depth of responses, potentially constraining the richness of the data gathered. 3. The absence of a specific definition for "data analytics" in the paper could have introduced variations in interpretation among survey respondents. 4. The study's findings may not be readily transferable to other countries or regions, given its focus on the Irish context. 5. The paper does not detail how well different data analysis methods work at finding financial statement fraud, which could be seen as a weakness. Fighting Accounting Fraud Through Forensic Data Analytics 1. The chosen accounting fraud cases may not be typical of all companies that file false financial reports. Also, companies not guilty of fraud are considered honest until proven otherwise. 2. Non-public companies were excluded from the study, given that the SEC's jurisdiction is limited to publicly traded entities. 3. The study recognizes accounting fraud's dynamic and evolving nature, suggesting that alternative information sources might lead to different conclusions. 4. Model performance in specific scenarios is suboptimal due to sample size limitations and the exclusion of

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points predictive variables. 5. To help people understand accounting fraud better, the study suggests adding more data, like qualitative variables, corporate governance data, information on insider trading, time-evolving features, and industry benchmarks. 6. Support vector machines, Bayesian models, and neural networks are recommended to improve the identification of fraudulent firms. 7. To look into and study patterns of accounting fraud that are unique to specific industries, it is suggested that the proposed method be used in other economic areas. Research Questions: Based on the GAP analysis, three research questions from the literature review of predictive analytics and accounting fraud detection: 1. To what extent can data analytics and predictive analytics enhance the accuracy and timeliness of fraud detection in financial audits? This research question delves into the effectiveness of data-driven auditing approaches in identifying fraudulent activities compared to traditional audit methods. Aim of this study is to find out if predictive models and data analytics techniques can help make fraud detection more accurate and audit turnaround times shorter by looking at how well they work. 2. How can data analytics effectively identify and classify different types of accounting fraud? This research question looks into how well data analytics and machine learning can tell the difference between different types of accounting fraud, like theft of assets, false financial reporting, and corruption. By analyzing financial data and identifying patterns associated with specific fraud types, this research aims to develop more targeted and effective fraud detection models. 3. What are the key challenges and obstacles to implementing data-driven auditing and fraud detection techniques in organizations? This research question investigates the obstacles organizations face in adopting data-driven auditing practices. This study tries to determine what strategies and resources are needed to make

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points data-driven fraud detection systems work by looking at costs, data availability, and the amount of knowledge that can be used. Problem/purpose statement The research investigates the transformative impact of predictive analytics on accounting fraud detection. The study aims to assess the practical implications, limitations, and overall effectiveness of implementing predictive analytics in organizations to identify and prevent fraudulent activities proactively. By addressing these questions, the research endeavors to contribute valuable insights into adopting and integrating predictive analytics as a strategic tool for countering the pervasive threat of accounting fraud. The study is motivated by the need to bridge existing gaps in understanding the nuances of predictive analytics application in fraud detection and its potential to revolutionize traditional approaches. This research is vital as it aims to provide a comprehensive understanding of the impact, challenges, and benefits associated with integrating predictive analytics into organizations' fraud detection mechanisms, ultimately contributing to advancing knowledge in the field [1,14]. The increasing prevalence and complexity of accounting fraud pose significant challenges for auditors and organizations in safeguarding financial integrity. Traditional audit methods, often relying on manual review and sampling, are becoming less effective in detecting sophisticated fraud schemes. Data-driven auditing, leveraging advanced analytics and machine learning, offers a promising approach to enhance fraud detection accuracy and timeliness. However, implementing these techniques faces several challenges, including data availability, expertise limitations, and integration with existing audit processes [1,16]. This study addresses the research topics to offer significant insights and recommendations for improving fraud detection capabilities and fostering financial transparency. This project aims to tackle the pressing requirement for enhanced fraud detection techniques by investigating the capabilities of data analytics and machine learning within the realm of financial audits. Addressed GAP From the research gap discussed above, below gap identified for this research: 1. The proposed approach focuses on analyzing annual reports, but there is a gap in understanding the integration of other data sources such as social media, news articles, or external audits. This aligns with the objective of assessing the practical implications and challenges associated with implementing predictive analytics. 2. The paper discusses the significance of big data analytics but needs more detailed information on techniques or algorithms. This aligns with the intention to evaluate the effectiveness of predictive analytics.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points 3. Experimental results show high accuracy, but there needs to be more understanding of the dataset's size and diversity. This aligns with assessing the practical implications and challenges. 4. The limited literature survey indicates a need for a more comprehensive analysis aligned to provide actionable recommendations. The gaps that have been detected in the articles are in line with the research objectives and the overarching problem statement. Our main goal with this study is to contribute to the field by filling in knowledge gaps and giving a complete picture of how crucial predictive analytics are for finding accounting fraud [7,11]. The objectives provide a framework for investigating the practical implications, problems, and transformative elements of adopting predictive analytics. They also ensure that the discovered gaps are consistently addressed. …………………………………………………………………………………………………… 6. Identify and write the research objectives (4-5 objectives) using the problem/purpose statement. The research objectives are concise, specific, and measurable. Use Bloom’s Taxonomy to identify appropriate active verbs for your research objectives. (25 points) The following is an example: My research project will:  Identify variables that have a statistically significant role in student drop out in online graduate courses.  Build a predictive model to identify at-risk students in online graduate courses.  Evaluate machine learning techniques for binary classification based on accuracy Research Objectives: In order to facilitate the research questions, the following objectives of the study have been drawn up using Bloom Taxonomy [12]: The research project will: 1. Evaluate the effectiveness of predictive analytics in early detection and prevention of accounting fraud. 2. Examine at the real-world effects, difficulties, and limits of using predictive analytics in computer systems that find business fraud. 3. Investigate how the adoption of predictive analytics transforms the traditional reactive approach to fraud prevention into a proactive, data-driven strategy.

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points 4. Assess how adding different types of data, like annual reports, social media, news articles, and external audits, affects the accuracy of predictive analytics in finding accounting fraud. 5. Provide actionable recommendations for organizations that adopt predictive analytics, addressing the identified gaps and challenges to enhance fraud detection strategies. …………………………………………………………………………………………………… 7. What are the ethical implications of your research? How do you plan to address them? The ethical issues may appear in any phase of your research including data collection, data management, data analysis, conclusions, and dissemination of findings. (25 points) The "Predictive Analytics in Accounting Fraud Detection" study encompasses various ethical issues that necessitate meticulous deliberation. The present discourse aims to delineate some significant ethical implications, accompanied by effective matching strategies to mitigate them. Privacy Concerns: Privacy concerns arise in predictive analytics, which frequently entails the examination of extensive datasets that may encompass confidential financial data. Privacy concerns may be raised concerning safeguarding individuals' financial data confidentiality. To safeguard individual identities, it is imperative to employ practical approaches for data anonymization and aggregation. It is imperative to rigorously adhere to data privacy standards and ensure express consent is obtained when required [3,18]. Bias and Fairness: Predictive models can unintentionally integrate biases that exist in past data, resulting in the possibility of unfair treatment or discrimination against specific individuals or groups. It is imperative to emphasize fairness during the model creation process by consistently assessing and mitigating any potential biases. In order to maintain fairness and impartiality, it is imperative to employ a variety of datasets and consistently update the model [3,18]. Transparency and Explainability: Predictive analytics models need to be straightforward and easy to understand because their complexity makes it hard for stakeholders to understand how decisions are made. The absence of transparency can potentially result in the development of mistrust. In order to build transparent models, it is essential to carefully record the methods used and make sure that the decision- making process is easy to understand. Deliver comprehensive elucidations of model outputs to pertinent parties [3,18]. Security Risks:

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points Due to the inherent sensitivity of financial data, there is a potential for security breaches or unauthorized access to the predictive analytics system. Incorporate resilient cybersecurity protocols encompassing encryption, stringent access limits, and periodic security audits. It is imperative to adhere to established industry norms of data security in order to mitigate potential risks and protect against vulnerabilities effectively [3,18]. Accountability and Responsibility: The issue of accountability and responsibility arises when a predictive analytics model produces erroneous or biased decisions, leading to difficulties in assigning blame and determining culpability. Establishing a comprehensive delineation of roles and duties within the research team is imperative. Develop protocols to ensure continuous monitoring and evaluation of the model's performance. This research aims to develop a methodical technique for quickly identifying and fixing issues [3,18]. Furthermore, it is imperative to form an ethical review board or seek help from established institutional review boards (IRB) to verify that the research adheres to ethical standards and values. Regular ethical training for the research team can augment their awareness and sensitivity toward potential ethical difficulties that may arise throughout the research endeavor. Consistently reviewing and revising these strategies in light of evolving ethical considerations will enhance the continuous ethical implementation of the research [3,18]. Bibliography [1] Ahmed Aboud and Barry Robinson. 2022. Fraudulent financial reporting and data analytics: an explanatory study from Ireland. Account. Res. J. 35, 1 (2022), 21–36. DOI:https://doi.org/10.1108/ARJ-04-2020-0079 [2] Oluwatoyin Esther Akinbowale, Polly Mashigo, and Mulatu Fekadu Zerihun. 2023. The integration of forensic accounting and big data technology frameworks for internal fraud mitigation in the banking industry. Cogent Bus. Manag. 10, 1 (2023), 1–23. DOI:https://doi.org/10.1080/23311975.2022.2163560 [3] Raul Artal and Sheldon Rubenfeld. 2017. Ethical issues in research. Best Pract. Res. Clin. Obstet. Gynaecol. 43, (2017), 107–114. DOI:https://doi.org/10.1016/j.bpobgyn.2016.12.006 [4] Deloitte. 2021. Predictive analytics for fraud detection . Retrieved from https://www2.deloitte.com/content/dam/Deloitte/tr/Documents/deloitte-analytics/tr-fraud- analytics.pdf [5] Luminița Ionescu. 2022. Machine Learning-based Decision-Making Systems, Cloud Computing and Blockchain Technologies, and Big Data Analytics Algorithms in Accounting and Auditing. Econ. Manag. Financ. Mark. 17, 4 (2022), 9.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

CIS 600: Research Methods Individual Assignment 2 Fall 2023 Due Date: Sunday, November 12, 11:59 PM 200 Points DOI:https://doi.org/10.22381/emfm17420221 [6] Yusuf Işık, İlker Kefe, and Jale Sağlar. 2023. Detection of fraudulent transactions using artificial neural networks and decision tree methods. Bus. Manag. Stud. An Int. J. 11, 2 (2023), 451–467. DOI:https://doi.org/10.15295/bmij.v11i2.2200 [7] Maria Jofre and Richard H. Gerlach. 2018. Fighting Accounting Fraud Through Forensic Data Analytics. SSRN Electron. J. (2018). DOI:https://doi.org/10.2139/ssrn.3176288 [8] KPMG. 2016. Using analytics successfully to detect fraud. 4 (2016), 5. Retrieved from http://www.blog.kpmgafrica.com/using-analytics-successfully-to-detect-fraud/? utm_campaign=57f76d81d4dbac5cf700744e&utm_content=5811e771d4dbac64a60061ce &utm_medium=smarpshare&utm_source=linkedin%5Cnhttps://assets.kpmg.com/ content/dam/kpmg/pdf/2016/07/using [9] Eman Nabrawi and Abdullah Alanazi. 2023. Fraud Detection in Healthcare Insurance Claims Using Machine Learning. Risks 11, 9 (2023), 160. DOI:https://doi.org/10.3390/risks11090160 [10] E. W.T. Ngai, Yong Hu, Y. H. Wong, Yijun Chen, and Xin Sun. 2011. The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature. Decis. Support Syst. 50, 3 (2011), 559–569. DOI:https://doi.org/10.1016/j.dss.2010.08.006 [11] Johan L. Perols, Robert M. Bowen, Carsten Zimmermann, and Basamba Samba. 2017. Finding needles in a haystack: Using data analytics to improve fraud prediction. Account. Rev. 92, 2 (2017), 221–245. DOI:https://doi.org/10.2308/accr-51562 [12] Theory Practice and Revising Bloom. 2008. A Revision of Bloom ’ s Taxonomy : An Overview David R . Krathwohl. ReVision 41, 4 (2008), 212–218. DOI:https://doi.org/10.1207/s15430421tip4104 [13] Pwc. 2020. Combat fraud & economic crime. Retrieved from https://www.pwc.com/gx/en/services/forensics/fraud.html [14] G. Sudha Sadasivam, Mutyala Subrahmanyam, Dasaraju Himachalam, Bhanu Prasad Pinnamaneni, and S. Maha Lakshme. 2016. Corporate governance fraud detection from annual reports using big data analytics. Int. J. Big Data Intell. 3, 1 (2016), 51. DOI:https://doi.org/10.1504/ijbdi.2016.073895 [15] Jay N. Shah. 2015. ‘Writing good effective title for journal article.’ J. Patan Acad. Heal. Sci. 1, 2 (2015), 1–3. DOI:https://doi.org/10.3126/jpahs.v1i2.16635 [16] Nitin Singh, Kee hung Lai, Markus Vejvar, and T. C.Edwin Cheng. 2019. Data-driven auditing: A predictive modeling approach to fraud detection and classification. J. Corp. Account. Financ. 30, 3 (2019), 64–82. DOI:https://doi.org/10.1002/jcaf.22389 [17] Milind Tullu. 2019. Writing the title and abstract for a research paper: Being concise, precise, and meticulous is the key. Saudi J. Anaesth. 13, 5 (2019), S12–S17. DOI:https://doi.org/10.4103/sja.SJA_685_18 [18] Camille Yip, Nian Lin Reena Han, and Ban Leong Sng. 2016. Legal and ethical issues in research. Indian J. Anaesth. 60, 9 (2016), 684–688. DOI:https://doi.org/10.4103/0019- 5049.190627

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

LR

Related Documents