Knowledge base

KB consists of :

• Taxonomy (terms & their relationships) in form of ontology
• Rule base
• Mathmatical model bases

Authors’ contributions

Roles: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Roles/Writing – original draft; Writing – review & editing.

Statistical model vs machine learning model

• sm can be interpreted by human e.g. doctors interpret arima models. Statistical modeling places a heavy emphasis on a model's underlying assumptions (e.g., linearity, normality of residuals, homoscedasticity). Choosing wrong assumptions and a respective sm modeling method can invalidate the conclusions drawn from the model.
• ml model is too complicated to be interpreted by human. That's why we let machine create it and call ml model.
• Statistics draws population inferences from a sample, and machine learning finds generalizable predictive patterns. (https://www.nature.com/articles/nmeth.4642)
• Statistics is The mathematics of the collection, organization, and interpretation of numerical data, especially the analysis of population characteristics by inference from sampling. Cf. American Heritage 

• The major difference between machine learning and statistics is their purpose. Machine learning models are designed to make the most accurate predictions possible. Statistical models are designed for inference about the relationships between variables. (https://towardsdatascience.com/the-actual-difference-between-statistics-and-machine-learning-64b49f07ea3)

• Decision tree algorithms have been studied throughout machine learning and statistics as a nonparametric approach to data modeling (Breiman, et al., 1984; Quinlan, 1993). Decision tree methodology is often contrasted with classical parametric statistical methodology, which requires the formulation of an explicit probabilistic model of the data generation process. (Cf. M.J. 1994, A Statistical Approach to Decision Tree Modeling) เราสรุปว่า DT เป็น supervised ML model ที่ตีความได้เหมือน statistical model และตอนตีความแต่ละเส้นทางในต้นไม้อาจเห็นโอกาสเกิดปัญหาในอนาคต เช่น เข้าไกล้ leave node ที่เป็นโรคแล้วไหมจะได้ป้องกันเนิ่นๆ

• XAI (Explainable ai) : methods and techniques in the application of artificial intelligence technology (AI) such that the results of the solution can be understood by humans. It contrasts with the concept of the "black box" in machine learning where even its designers cannot explain why an AI arrived at a specific decision. Wikipedia; https://cacm.acm.org/magazines/2022/4/259398-explainable-ai/fulltext#

Bkk night light

OCR by Google Doc

https://www.dailygizmo.tv/2020/01/31/3-howto-pdf-to-taxt-on-google-drive/?fbclid=IwAR2iu8U17vpWJdv-PwBfsDOnv0_e-fCmela0IwJx6lLIb_z8es0dlOdzr0w

ล้มเร็วเรียนรู้เร็ว

เป็นหลักทางการตลาดว่าให้รีบลงมือทำจะมัวแต่วางแผนจะได้รู้ว่าหมู่หรือจ่า ถ้าล้มเหลวจะได้รีบเอาเวลาไปทำอย่างอื่นแทน

Generative adversarial network (GAN)

A generative adversarial network (GAN) is a class of machine learning systems where two neural networks contest with each other in a game (in the sense of game theory, often but not always in the form of a zero-sum game). Given a training set, GAN learns to generate new data with the same statistics as the training set. For example, a GAN trained on photographs can generate new photographs that look at least superficially authentic to human observers, having many realistic characteristics. Though originally proposed as a form of generative model for unsupervised learning, GANs have also proven useful for semi-supervised learning, fully supervised learning, and reinforcement learning.

https://medium.com/@nutorbitx/gans-%E0%B8%AD%E0%B8%B0%E0%B9%84%E0%B8%A3%E0%B8%84%E0%B8%B7%E0%B8%AD-generative-adversarial-networks-7973ae70db70

The well-known application of GAN is deepfake.

Parsimonious model

A parsimonious model is a model that accomplishes a desired level of explanation or prediction with as few predictor variables as possible.

Pseudonymization vs. Anonymization

Pseudonymization is a data management and de-identification procedure by which personally identifiable information fields within a data record are replaced by one or more artificial identifiers, or pseudonyms.

Anonymization is tokenization.

https://dataprivacymanager.net/pseudonymization-according-to-the-gdpr/

Google data studio

https://datastudio.google.com/u/0/navigation/reporting

Mashup

A mashup (computer industry jargon), in web development, is a web page or web application that uses content from more than one source to create a single new service displayed in a single graphical interface.

--https://en.wikipedia.org/wiki/Mashup_(web_application_hybrid)

Box plot

https://en.wikipedia.org/wiki/Box_plot

https://www.tableau.com/about/blog/2014/12/5-chart-types-youve-never-tried-tableau-35281

Graph DBMS

https://neo4j.com/

Example graph DBMS is https://neo4j.com/ which claims good performance (equivalent to RDBMS + indexing; the indexing is replaced with graph edges in neo4j).

Java APIs

Publicly available ISO standards

https://standards.iso.org/ittf/PubliclyAvailableStandards/index.html

Handle missing values in ML

https://towardsdatascience.com/how-to-handle-missing-data-8646b18db0d4

VA vs. W

VA คือกำลังไฟสกปรก + กำลังไฟสะอาด (W)