Python 3.6 Jupyter Notebook

Living labs

Your completion of the notebook exercises will be graded based on your ability to do the following:

Understand: Do your comments show evidence that you recall and understand technical concepts?

Notebook objectives:

By the end of this notebook, you will be expected to understand the use and opportunities provided by the living labs approach.

List of exercises:

  • Exercise 1: Provide advantages of living labs and deep data.
  • Exercise 2: List the roles of the different stakeholders in a living labs setup.
  • Exercise 3: Explain the change of the role of a data analyst into a more strategic position in the modern organization.

Notebook introduction

Living labs describe the paradigm of working with new ideas and technology, and directly engaging with and observing users while they are living their lives.

While the levels of direct user engagement and co-creation vary between the examples referenced in this section, they share the access to users’ behavior in response to novel products, content, or activity.

1. Living labs

A living lab can be established using existing infrastructure and data sources. This was shown in the example of the Andorra living lab (discussed in the video content) which made use of CDRs, credit card transactions, and public transportation data. While direct user engagement may be limited, their behaviors – especially their reactions to new products or experimental interventions – can be observed in this setting. Living labs can also be entirely virtual, as is the case in the A/B testing of web applications.

Although the bleeding edge of living lab development is driven by the growing data collection and interaction capabilities that have been enabled by the spread of ubiquitous computing, this is not a prerequisite for big or electronic data. In the video content, Professor Alex Pentland discusses an example of how Nike leverages living labs to determine which shoes to release within their stores. You can also refer to Hack-MIT as an example of how to use existing infrastructure, such as WiFi access points, as a simple living lab.

Data visualization can take many forms. You can refer to an example of a visualization of WiFi data. Once you have the basic lab in place, you would generally build your use case around it. Marketers would likely be interested in the density and profiles of individuals in specific areas, and the resulting efficiency of campaigns or other interventions, while city planners may be interested in optimizing flow in public spaces.

All of the data that was used in this course was generated by sophisticated deployments that collected not only big data, but also deep data.

Another recent trend in technology, the "Internet of Things" (or "IoT"), holds significant opportunities for data collection and interaction with the environment, whether human or device based. The big shift here is that the end-points are becoming active and they contain computation capabilities, where previous efforts focused on sensors or observations alone.

Note: The presence of computation capabilities at the endpoints also opens up opportunities such as software-defined products. Consider activity trackers learning to recognize new activities, or self-driving cars where new features (such as parking capabilities) can be added with software updates. Once you have gathered data and refined your algorithms, these can potentially be implemented as software-defined products or features (like the one above), and can be added through software updates.

You can read more about the European open living labs network, and another example of a recent deployment of a living lab using sensor and mobile data, the Amsterdam IoT living lab. Additional information on the Amsterdam living lab can be found through this wiki, as well as a blog post about building the world’s biggest iBeacon living lab.

Note:

Living lab projects typically include a wide variety of stakeholders and partners, including government, academic, and commercial parties. Refer to the goal statement of the Amsterdam Smart City project (2016) below as an example:

“The goal of the IoT Living Lab is to provide IoT infrastructure and actionable, Open Data, and developer friendly platforms for emerging IoT innovations. This stimulates the creation of new startups and mobile applications, which in turn make a rapid impact on the local economy.”

(Amsterdam Smart City 2016)

While the focus of this course is on social analytics, there are a number of recent technological trends that can add significant value to your future projects. You are encouraged to explore these on your own.

Much of the publicity around big data focuses solely on the volume or, in some cases, the format of the data, and many fail to capitalize on existing sources of data that may already be accessible. Dark data refers to data that is already available, but not utilized. IoT also contains a number of relevant concepts. You can read about Gartner's view of the top 10 IoT technologies for 2017 and 2018 for more information on this topic.

Note:

Once you better understand the available data sources, as well as the options offered by technological developments, you will be in a much better position to successfully start and complete your social analytics projects.

Refer to the additional links below for more guidance:

Exercise 1 Start.

Instructions

In the video content, Arek Stopczynski points out the dangers of fixating on big data and losing perspective, as well as studying the data instead of the population of interest (and interactions within).

a) How do "deep data" and "living labs" help in avoiding the pitfalls of only studying available data sets?

b) List some of the advantages of "living labs" (as opposed to "focus groups" or "large scale surveys").

Note: Your answer should be a short submission (two to three sentences) for each of the two questions.

Your markdown answer here.


In [ ]:


Exercise 1 End.

Exercise complete:

This is a good time to "Save and Checkpoint".

1.1 Purpose

Typical purposes for setting up living labs include:

  • Research;
  • Development; and
  • Production applications.

In a commercial context, there are internal and external opportunities for setting up living labs, which are covered in the following sections.

1.1.1 External

Typical uses of living labs include marketing and customer insight use cases where the profile and demographics of individuals are used to optimize or create product or service offerings. Usually, these insights are also relevant in supply chain optimization.

1.1.2 Internal

A better understanding of social networks in companies can be used in human resource (HR) projects, as per the examples introduced in Module 7. You can also refer to the work of Daniel Olguin Olguin et al. (2009), whose Sociometric Badge study provides a myriad of insights into this topic. The paper titled “Sensible Organizations: Technology and Methodology for Automatically Measuring Organizational Behavior” is freely available.

Organizations usually have access to large amounts of dark and deep data, which you can use to get started. These sources of data can also be extended with applications (such as Funf) to create deep data sets.

A better understanding of social networks in companies can be used in human resources projects, as per the examples introduced in Module 7. You can also refer to the seminal work of Professor Alex Pentland, the Sociometric Badge study from 2008, which provides a myriad of insights in this paper.

Note:

It is hard to overemphasize the importance of the data privacy, or even data sovereignty, of individuals due to its centrality in building the trust relationship necessary for a living lab to be successful. Review the course content on privacy (Module 6), with special focus on the open Personal Data Store (openPDS) architecture, which strives to provide privacy to the users even when the data is used for internal purposes only.

Exercise 2 Start.

Instructions

In the video content, David Shrier discusses the typical stakeholders involved in setting up living labs. Provide a short summary of the roles played by each of the following parties, referencing both the input (what they require from the other parties) and the output that they deliver to the other stakeholders.

a) Government

b) Data partners

c) Local or global business

d) Local universities or academic partners

Your markdown answer here.


In [ ]:


Exercise 2 End.

Exercise complete:

This is a good time to "Save and Checkpoint".

Exercise 3 Start.

Instructions

In the video content, Professor Alex Pentland discussed how many organizations have begun to realize the value of data as a strategic asset. Keeping this in mind, consider the following question:

  • How does the role of an analyst change when considering strategic analysis as opposed to more traditional data analysis?

Hint:

Provide a short description of the typical tasks that you would expect within this new role, and briefly discuss or refer to potential organizational changes or parties that the analysts would need to implement or interact with in their new role.

Your markdown answer here.


In [ ]:


Exercise 3 End.

Exercise complete:

This is a good time to "Save and Checkpoint".

2. Submit your notebook

Please make sure that you:

  • Perform a final "Save and Checkpoint";
  • Download a copy of the notebook in ".ipynb" format to your local machine using "File", "Download as", and "IPython Notebook (.ipynb)"; and
  • Submit a copy of this file to the Online Campus.

3. References

Amsterdam Smart City. 2016. “IoT Living Lab - Amsterdam Smart City.” Accessed October 9. https://amsterdamsmartcity.com/projects/iot-living-lab.

Note:

Arek Stopczynski references the Copenhagen Network Study and indicates that the research is ongoing. You can read more about recent developments and additional publications on Sune Lehmann’s website.


In [ ]: