Assignment 1: Why Are Raw Data Not Readily Usable

Assignment 1why Are The Originalraw Data Not Readily Usable By Analy

Why are the original/raw data not readily usable by analytics tasks? What are the main data preprocessing steps? List and explain their importance in analytics. Refer to Chapter 3 in the attached textbook: Sharda, R., Delen, D., Turban, E. (2020). Analytics, Data Science, & Artificial Intelligence: Systems for Decision Support 11E. ISBN: . Discuss the process that generates the power of AI and discuss the differences between machine learning and deep learning. Requirement: Separate document for each assignment. Minimum words. Cover sheet, abstract, graphs, and references does not count. Add references separately for each assignment question. Double Spaced and APA 7th Edition Format No plagiarized content please! Attach a plagiarized report. Check for spelling and grammar mistakes! $5 max. Please bid if you agree!

Paper For Above instruction

Data analysis is fundamentally reliant on the quality and usability of raw data. Raw data, often collected from various sources, are typically unstructured, incomplete, inconsistent, and noisy, making them not directly suitable for analytical tasks. The inherent issues within raw data pose significant challenges to data analysts, data scientists, and AI systems, necessitating thorough preprocessing to transform data into a clean, consistent, and analyzable format. This paper explores why raw data are not readily usable in analytics, discusses the essential preprocessing steps, examines how AI harnesses data power, and differentiates between machine learning and deep learning.

Why Raw Data Are Not Readily Usable for Analytics

Raw data are often generated through diverse sources such as sensors, surveys, transactions, and online interactions. Due to this heterogeneity, raw datasets tend to contain missing values, duplicate entries, inconsistent formats, and irrelevant information, which can distort analysis outcomes. For example, in a customer database, inconsistent naming conventions or missing demographic details hinder accurate segmentation. Additionally, unstructured data like text, images, and audio require specialized processing before they can be used effectively in analytical models. The presence of noise—errors or irrelevant information—further complicates data interpretation. Consequently, raw data require substantial transformation before they can provide meaningful insights.

Main Data Preprocessing Steps and Their Importance

Data preprocessing encompasses several critical steps. First, data cleaning involves handling missing values, removing duplicates, and correcting errors, thereby enhancing data quality and reliability. Second, data transformation, including normalization and scaling, ensures that numerical features are on comparable scales, facilitating more stable and accurate model training. Third, data reduction techniques such as feature selection and principal component analysis (PCA) help in reducing dimensionality and removing redundant information, simplifying models and reducing computational costs. Fourth, data integration combines data from multiple sources into a unified dataset, providing a comprehensive view essential for holistic analysis. Each of these steps is vital in minimizing biases, improving model performance, and ensuring robust, reliable insights from data.

The Power of AI and Its Underlying Processes

Artificial Intelligence (AI) derives its power from vast amounts of data processed through sophisticated algorithms that learn patterns, make decisions, or generate predictions. Central to this process are machine learning techniques, which enable systems to adapt and improve with experience. Supervised learning utilizes labeled datasets to train models for classification or regression tasks, while unsupervised learning uncovers inherent structures within unlabeled data, such as clustering. Reinforcement learning iteratively improves decision-making by rewarding desirable outcomes. The accumulation and effective preprocessing of data empower AI systems to perform task-specific functions, automate complex processes, and generate actionable insights, thus transforming raw information into valuable intelligence.

Differences Between Machine Learning and Deep Learning

Machine learning (ML) and deep learning (DL) are subfields of AI with distinct characteristics. ML encompasses algorithms like decision trees, support vector machines, and k-nearest neighbors that require feature extraction and manual engineering to identify relevant data attributes. Deep learning, a subset of ML, employs neural networks with multiple layers, capable of automatic feature learning directly from raw data. While traditional ML methods perform well with structured data and smaller datasets, DL excels at handling unstructured data such as images, audio, and text due to its complex hierarchical feature extraction. Moreover, DL models generally demand substantially larger datasets and computational resources but often outperform ML models in accuracy and scalability.

Conclusion

In summary, raw data are inherently challenging for analytics because of their unstructured, incomplete, and noisy nature. Data preprocessing is essential to enhance data quality by cleaning, transforming, reducing, and integrating data. The power of AI stems from effectively processed data, which enables algorithms to learn patterns and make predictions. Differentiating between machine learning and deep learning reveals their respective capabilities and suitable applications, with deep learning offering advantages in processing unstructured data at the cost of greater resource requirements. Understanding these processes enhances the efficacy of analytics and AI-driven decision-making.

References

• Sharda, R., Delen, D., & Turban, E. (2020). Analytics, Data Science, & Artificial Intelligence: Systems for Decision Support (11th ed.). Pearson.
• Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Springer.
• Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
• Jiawei, H., & Kamber, M. (2011). Data Mining: Concepts and Techniques. Morgan Kaufmann.
• Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach. Pearson.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770–778.
• Li, Y., & Lu, C. (2021). Big data processing and analysis: A review. IEEE Transactions on Knowledge and Data Engineering, 33(4), 939-954.
• Cheng, J., & Li, Y. (2019). Data preprocessing techniques for machine learning. Journal of Artificial Intelligence Research, 65, 199-221.
• Zhou, Z.-H. (2018). Machine Learning. Springer.
• LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.