Showing posts with label metrica. Show all posts
Showing posts with label metrica. Show all posts

30 November 2018

🔭Data Science: Metrics (Just the Quotes)

"Little experience is sufficient to show that the traditional machinery of statistical processes is wholly unsuited to the needs of practical research. Not only does it take a cannon to shoot a sparrow, but it misses the sparrow! The elaborate mechanism built on the theory of infinitely large samples is not accurate enough for simple laboratory data. Only by systematically tackling small sample problems on their metrics does it seem possible to apply accurate tests to practical data." (Ronald A Fisher,Statistical Methods for Research Workers", 1925), 

"The degree of confirmation assigned to any given hypothesis is sensitive to properties of the entire belief system [...] simplicity, plausibility, and conservatism are properties that theories have in virtue of their relation to the whole structure of scientific beliefs taken collectively. A measure of conservatism or simplicity would be a metric over global properties of belief systems." (Jerry Fodor,Modularity of Mind", 1983)

"No subjective metric can escape strategic gaming [...] The possibility of mischief is bottomless. Fighting ratings is fruitless, as they satisfy a very human need. If one scheme is beaten down, another will take its place and wear its flaws. Big Data just deepens the danger. The more complex the rating formulas, the more numerous the opportunities there are to dress up the numbers. The larger the data sets, the harder it is to audit them." (Kaiser Fung,Numbersense: How To Use Big Data To Your Advantage", 2013)

"The urge to tinker with a formula is a hunger that keeps coming back. Tinkering almost always leads to more complexity. The more complicated the metric, the harder it is for users to learn how to affect the metric, and the less likely it is to improve it." (Kaiser Fung," Numbersense: How To Use Big Data To Your Advantage", 2013)

"Precision and recall are ways of monitoring the power of the machine learning implementation. Precision is a metric that monitors the percentage of true positives. […] Recall is the ratio of true positives to true positive plus false negatives." (Matthew Kirk,"Thoughtful Machine Learning", 2015)

"Just as with aggregated data, an average is a summary statistic that can tell you something about the data - but it is only one metric, and oftentimes a deceiving one at that. By taking all of the data and boiling it down to one value, an average" (and other summary statistics) may imply that all of the underlying data is the same, even when it’s not." (John H Johnson & Mike Gluck,Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Roughly stated, the No Free Lunch theorem states that in the lack of prior knowledge" (i.e. inductive bias) on average all predictive algorithms that search for the minimum classification error" (or extremum over any risk metric) have identical performance according to any measure." (N D Lewis,Deep Learning Made Easy with R: A Gentle Introduction for Data Science", 2016)

"Variables which follow symmetric, bell-shaped distributions tend to be nice as features in models. They show substantial variation, so they can be used to discriminate between things, but not over such a wide range that outliers are overwhelming." (Steven S Skiena,The Data Science Design Manual", 2017)

"Dimensionality reduction is a way of reducing a large number of different measures into a smaller set of metrics. The intent is that the reduced metrics are a simpler description of the complex space that retains most of the meaning." (Danyel Fisher & Miriah Meyer,Making Data Visual", 2018)

"Statistical metrics can show us facts and trends that would be impossible to see in any other way, but often they’re used as a substitute for relevant experience, by managers or politicians without specific expertise or a close-up view." (Tim Harford,The Data Detective: Ten easy rules to make sense of statistics", 2020)

"Non-linear associations are also quantifiable. Even linear regression can be used to model some non-linear relationships. This is possible because linear regression has to be linear in parameters, not necessarily in the data. More complex relationships can be quantified using entropy-based metrics such as mutual information. Linear models can also handle interaction terms. We talk about interaction when the model’s output depends on a multiplicative relationship between two or more variables." (Aleksander Molak,Causal Inference and Discovery in Python", 2023)

"Defining an appropriate distance metric from the context ofthe problem is a most important decision. For example, if your variables are all numeric, and on the same scale, then Euclidean distance might be best. If your variables are categorical, you might need to use something like Hamming distance." (Dianne Cook & Ursula Laa, "Interactively Exploring High-Dimensional Data and Models in R", 2026)

"Technical features are the design and training decisions that are made that determine aspects such as a model’s use-case suitability and performance metrics. By understanding the basics of these technical features, you can better evaluate, assess, and predict how models will perform. These technical features include architecture, parameter count, training data, fine-tuning and specialized tuning, context window, and unique functionalities." (Chris Minnick, "A Developer’s Guide to Integrating Generative AI into Applications", 2026) 

"When the ratio of the largest to smallest observation is large you should question whether the data are being analyzed in the right metric" (transformation)." (George E P Box)

30 November 2015

🪙Business Intelligence: Metrics (Just the Quotes)

"Although it does so much more, the central purpose of a dashboard is to warn the user when any relevant metrics are out of acceptable boundaries. In the dashboard terminology, these alerts consisting of rules and actions add critical value to an enterprise dashboard deployment complemented with strong visual indicators of warnings." (Shadan Malik, "Enterprise Dashboards: Design and best practices for IT", 2005)

"Metrics are measurements of activities to evaluate performance, mostly within a relative framework of time, geography, and aggregation." (Shadan Malik, "Enterprise Dashboards: Design and best practices for IT", 2005)

"The distinguishing feature is that a dashboard is an application with a collection of metrics, benchmarks, goals, results, and alerts presented in a visually effective manner, whereas a portal is a collection of different applications presented together within a personalized framework." (Shadan Malik, "Enterprise Dashboards: Design and best practices for IT", 2005)

"[…] a KPI is a metric, but a metric is not necessarily a KPI. A metric is really a measure of anything. A KPI, however, is meant to be a measure “ that matters ”and that ideally can be acted on. An organization has many metrics, but typically just a few KPIs. This is often the challenge when a management team meets to agree on a short list of key measures for use in its proposed dashboards and scorecards." (Nils H Rasmussen et al, "Business Dashboards: A visual catalog for design and deployment", 2009)

"A culture of data fluency needs to be built on a shared understanding of the data sources, data analysis, key metrics, and data products. It requires employees to be on the same page about how data is used and why it is important." (Zach Gemignani et al, "Data Fluency", 2014)

"Any presentation of data, whether a simple calculated metric or a complex predictive model, is going to have a set of assumptions and choices that the producer has made to get to the output. The more that these can be made explicit, the more the audience of the data will be open to accepting the message offered by the presenter." (Zach Gemignani et al, "Data Fluency", 2014)

"Goals associated with a few, well-understood key metrics is a powerful combination. For both internal and external stakeholders, there is a strong alignment between organization mission, vision, goals, and tracking of progress. The efforts of everyone can be directed at these measurable goals, and people will focus on the processes that can impact these metrics." (Zach Gemignani et al, "Data Fluency", 2014)

"Metrics can serve two purposes: identifying problems and measuring performance. When the goal is to identify problems and pinpoint areas of operational inefficiency and ineffectiveness, defining the right metric requires a bit of detective work. It requires you to uncover the data residue of a problem and to determine what evidence can be found and how exactly it shows up. When the goal is to measure performance, the right success metrics focus on measures that can be controlled and where improvement in the metric is an unambiguously good thing." (Zach Gemignani et al, "Data Fluency", 2014)

"[…] an overall green status indicator doesn’t mean anything most of the time. All it says is that the things under measurement seem okay. But there always will be many more things not under measurement. To celebrate green indicators is to ignore the unknowns. […] The tendency to roll up metrics into dashboards promotes ignorance of the real situation on the ground. We forget that we only see what is under measurement. We only act when something is not green." (Sriram Narayan, "Agile IT Organization Design: For Digital Transformation and Continuous Delivery", 2015)

"Grouping a bunch of related fine-grained metrics into an aggregate metric works better than handling them individually. This is commonly achieved by defining an aggregate metric as a weighted sum of contributory metrics. It is the sort of technique used in credit scoring, insurance risk scoring, or the points system for immigration. Thresholds are then set on the aggregate score to define different categories of eligibility." (Sriram Narayan, "Agile IT Organization Design: For Digital Transformation and Continuous Delivery", 2015)

"Rolling up fine-grained metrics to create high-level dashboards puts pressure on teams to keep the fine-grained metrics green even when it might not be the best use of their time." (Sriram Narayan, "Agile IT Organization Design: For Digital Transformation and Continuous Delivery", 2015)

"Scaling supervision using metrics is one thing; scaling results is quite another. The former doesn’t automatically ensure the latter." (Sriram Narayan, "Agile IT Organization Design: For Digital Transformation and Continuous Delivery", 2015)

"Simplicity for data visualization often focuses on minimizing the number of elements that do not add value to your display. These include borders, gridlines, axes lines, and boxes, which can easily distract from your core message. This recommendation also relates to the information itself. You should strive to create a visualization that focuses on specific aspects of the data, rather than including all fields and metrics but not saying much about any of them." (Andy Kriebel & Eva Murray, "#MakeoverMonday: Improving How We Visualize and Analyze Data, One Chart at a Time", 2018)

"The field of [data] visualization takes on that goal more broadly: rather than attempting to identify a single metric, the analyst instead tries to look more holistically across the data to get a usable, actionable answer. Arriving at that answer might involve exploring multiple attributes, and using a number of views that allow the ideas to come together. Thus, operationalization in the context of visualization is the process of identifying tasks to be performed over the dataset that are a reasonable approximation of the high-level question of interest." (Danyel Fisher & Miriah Meyer, "Making Data Visual", 2018)

"Data observability platforms often consolidate various data health metrics, such as freshness, accuracy, completeness, and consistency, into a unified dashboard, offering a holistic view of data health. This may take the form of a data scorecard. These tools can also provide metrics such as system uptime/availability, data processing times, error rates in data processing, query performance metrics, resource utilization, and user engagement metrics. They provide customizable key performance indicators (KPIs), real-time alerts, and root cause analysis, and they can help assess the impact of data quality issues on downstream applications, reports, and business processes." (Fern Halper, "Data Makes the World Go 'Round", 2026)

Related Posts Plugin for WordPress, Blogger...

About Me

My photo
Koeln, NRW, Germany
IT Professional with more than 25 years experience in IT in the area of full life-cycle of Web/Desktop/Database Applications Development, Software Engineering, Consultancy, Data Management, Data Quality, Data Migrations, Reporting, ERP implementations & support, Team/Project/IT Management, etc.