這是誰的會議?

(從我的一個喜歡與尊敬的作家,賽斯 高汀

一個簡單的清單,一題一分。

你的下一次會議(不是會話,不是演講,還是會議)能通過這個測試嗎?

一)有一個人要負責。

二)只用分配的時間與所需的時間匹配,而不是日曆上所說的時間。

三)被邀請的每一個人都是需要在那裡,並且關鍵的人都在。

四)如果有人不來,那表示那個人的默認。

五)沒有比這個會議有更好的方式來推動這一進展。

六)明確陳述期望的結果。 組織者要描述“這就是我想要發生的事情”或是有“是的,我們就完成了”,那這個會議就可以結束了。

七)所有相關的信息與包括分析,都可以在會議之前有大量時間去進行審查。

如果你能得到全部七分,我會來參加。

Regression Machine Learning Template Using Python

As I work on practicing and solving machine learning (ML) problems, I find myself repeating a set of steps and activities repeatedly.

Thanks to Dr. Jason Brownlee’s suggestions on creating a machine learning template, I have pulled together a project template that can be used to support regression ML problems using Python.

The purpose of the template is to show the data preparation, ML modeling using algorithms, and performance tuning steps. You can download the template from the Machine Learning Project Templates page.

You can also check out the sample HTML-formatted report here on GitHub.

They Are Not Trying to be Creepy

In his podcast, Akimbo [https://www.akimbo.me/], Seth Godin teaches us how to adopt a posture of possibility, change the culture, and choose to make a difference. Here are my takeaways from the episode.

  • Our cultures and our lives are inextricably linked with media, and someone has pay for the media we consume. We can pay for the media in three ways: 1) pay for the media, 2) pay for the thing that comes within the media, or 3) an advertiser can pay for the media.
  • When an advertiser pays for a piece of media we consume, we are kidding ourselves if we pretend what the advertiser wants does not matter.
  • There are two types of marketing: brand marketing and direct marketing. Either approach has its purposes and goals.
  • Brand advertising has been the primary vehicle for marketing for a long time. This approach tries to influence our perception of a brand subtly
  • Direct marketing is all about actions and measurements. Will someone click or buy after seeing the direct advertisement? Even a 1% response rate is a home run.
  • Advertisers would prefer to buy the mass and reach everyone. That is difficult to do because the interests are so divergent on the net. Companies are trying hard to collect data from individuals in an attempt to find the micro markets that exist. If those micro markets can be found, the data can be sold to the advertisers.
  • Once the data gathering started, companies apply the direct marketing approach to leverage the data, and everyone does it. A downward ratchet started.
  • With all these direct marketing efforts needing the eyeballs and clicks, consumers become the product. The free services exist to sell “you” to the advertisers. We should be aware of this and know what we are getting ourselves into.
  • As we turn the brand advertisers into the direct marketers, we should be acutely aware of the societal and cultural implications. It is up to us on what we would tolerate and set up the guard rails.
  • Our opportunity her is not to race faster to the bottom. We need to be careful with how we spend our attention and on whom every day.
  • “Our culture defines who we are, is what we got, and what we make it into.”

Simple Classification Model for Bank Marketing Using Python

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

Dataset Used: Bank Marketing Dataset

Dataset ML Model: Binary classification with numerical and categorical attributes

Dataset Reference: http://archive.ics.uci.edu/ml/datasets/bank+marketing

One source of potential performance benchmarks: https://www.kaggle.com/rouseguy/bankbalanced

INTRODUCTION: The Bank Marketing dataset involves predicting the whether the bank clients will subscribe (yes/no) a term deposit (target variable). It is a binary (2-class) classification problem. There are over 45,000 observations with 16 input variables and 1 output variable. There are no missing values in the dataset.

CONCLUSION: The baseline performance of the 11 algorithms achieved an average accuracy of 89.13%. Three algorithms (Stochastic Gradient Boosting, Random Forest, and AdaBoost) achieved the top accuracy and Kappa scores. The top result achieved using the training data was from Stochastic Gradient Boosting. It achieved an average accuracy of 91.00% after a series of tuning trials, and its accuracy on processing the validation dataset was 90.58%. For this project, the Stochastic Gradient Boosting ensemble algorithm yielded consistently top-notch training and validation results, which warrant the additional processing required by the algorithm.

The HTML formatted report can be found here on GitHub.

什麼是和什麼可能是

(從我的一個喜歡與尊敬的作家,賽斯 高汀

他們的共同點比你想像的要少得多。

創造更美好未來的關鍵步驟就是堅持不以過去的假設,不滿,和死胡同來做為基礎。

未來將不會十全十美。我們也不會變成為完美。但是我們可以變的更善良。 我們可以去多聽。 我們甚至可以給予自己多幾個機會。

未來並不總是會成功。 我們不會總是能成功。 但我們可以保持警惕並尋求可能,而不只是聽從他人的預測。

未來並不總是公平的。 但我們可以嘗試,我們可以關心,我們可以選擇與他人共鳴與合作。

事情會變更好如果我們能給我們自己一個許可去做。

Stakeholder Analysis

In the book, Bare Bones Change Management: What you shouldn’t not do, Bob Lewis explained the seven must-have elements for any change management effort to have a chance of succeeding. Here are my takeaways from one of the topics discussed in the book.

Change efforts do not succeed just because they are right for the organization. They succeed because they benefit someone.

Return on Investment from the Cost-Benefit Analysis is often the invitation for a change to enter the organization.

There are three significant stakeholder groups: the Supporters, the Acceptors, and the Resistors. Think of the distribution as a bell curve. Wise change manager will always perform an in-depth analysis of the stakeholders.

Look for supporters among people who are ambitious, who complain about the old way of doing things, and who are confident in their abilities to adapt.

On the flipside, change manager can usually find resistors who expect to lose their standing and influence due to the change. They have built up skills and value which will be threatened by the change. The resistors can also be those who are less confident in their abilities to adapt than the supporters.

The acceptors, a majority in the organization, will not actively support or resist the change. Leadership is the best way to reach that population because that is what they look for. The direction can come from anyone, even from the resistors, so it is best the leadership comes from the change manager.

Bob further suggested some tactics to consider for managing the stakeholder groups. They include Making it a win, Involving, Promoting, Communicating, Introducing, Giving special treatment, Defusing, Cornering, and Marginalizing.

Managing the stakeholder groups is probably the most challenging element of leading the change effort.

Multi-Class Machine Learning Template Using Python

As I work on practicing and solving machine learning (ML) problems, I find myself repeating a set of steps and activities repeatedly.

Thanks to Dr. Jason Brownlee’s suggestions on creating a machine learning template, I have pulled together a project template that can be used to support multi-class ML problems using Python.

The purpose of the template is to show the data preparation, ML modeling using algorithms, and performance tuning steps. You can download the template from the Machine Learning Project Templates page.

You can also check out the sample HTML-formatted report here on GitHub.

當你的主意被盜用

(從我的一個尊敬的作家與英雄,賽斯 高汀

有一些沉思:

對你是有好處。當別人認為你的主意值得竊取,這不是更好嗎?如果你一直創作,創造出那本書或那部電影或是那個概念,但並沒有人想要貶低它,擴展它,或者與它一起運行,當你的主意是無人問津,這會更好嗎?

你的主意不會用盡。其實,當越多的人抓住你的主意並與他們一起製造新鮮事,你的發件箱裡就會出現越來越多的空間,這意味著你可以出產更多的主意,對吧?

當一個主意被發展,那就是好主意。他們會充實我們的文化,建立聯繫並改善我們的生活。 這不就是你為什麼要去創造你的主意,不是嗎?

有主意的目標不是爭取信用。而它的目標是改變我們的周圍。