**5.4 Method 3: crowdsourcing**

The second method involves a human-based approach, as a crowdsourcing process. In this method, the crowdsourcing process was achieved via a human-based outsourcing platform called "Amazon Mechanical Turk." The "Amazon Mechanical Turk" platform is a crowdsourcing Internet marketplace, operated by "Amazon," which enables individuals to coordinate the use of human intelligence and perform

**Figure 7.** *Oakland, walkability, and car dependency (figure was created by the author) [17].*

tasks that computers are currently unable to perform successfully. It is an on-demand sample of users that executes simple assignments over an agreed period of time. The "Amazon Mechanical Turk" can be associated with the term "Human Computing."

Initially, and in order to test the feasibility of this method, we processed only few blocks in Emeryville, Oakland, and the questions posted to Amazon Mechanical Turk were very simple and required identification of certain elements and whether they appear in the images. Example elements that were queried are bicycles, lofts, abandoned buildings, and industrial buildings (**Figure 11**). In the next stage, a large group was given two different sets of questions. The first set of questions is related to human subjectivity, which implies that the users were given a set of subjective questions that were related to the qualitative rating of selected neighborhoods in the San Francisco Bay Area. The questions were communicated in a simple way, by extracting Google Street viewpoints as images and submitting

**163**

**Figure 9.**

**Figure 8.**

*author) [23].*

*Human-Centered Approaches in Urban Analytics and Placemaking*

*Oakland crime reports, 2010 (left), Oakland crime reports 2013 (right) (figure was created by the* 

*Oakland, industrial buildings, lofts/luxury residence, businesses related to artists, yoga and fitness studios,* 

*fashionable cafes (figure was created by the author) [23–25].*

*DOI: http://dx.doi.org/10.5772/intechopen.89675*

*Human-Centered Approaches in Urban Analytics and Placemaking DOI: http://dx.doi.org/10.5772/intechopen.89675*

**Figure 8.**

*Sustainability in Urban Planning and Design*

tasks that computers are currently unable to perform successfully. It is an on-demand sample of users that executes simple assignments over an agreed period of time. The "Amazon Mechanical Turk" can be associated with the term "Human Computing." Initially, and in order to test the feasibility of this method, we processed only few blocks in Emeryville, Oakland, and the questions posted to Amazon Mechanical Turk were very simple and required identification of certain elements and whether they appear in the images. Example elements that were queried are bicycles, lofts, abandoned buildings, and industrial buildings (**Figure 11**). In the next stage, a large group was given two different sets of questions. The first set of questions is related to human subjectivity, which implies that the users were given a set of subjective questions that were related to the qualitative rating of selected neighborhoods in the San Francisco Bay Area. The questions were communicated in a simple way, by extracting Google Street viewpoints as images and submitting

*Oakland, walkability, and car dependency (figure was created by the author) [17].*

**162**

**Figure 7.**

*Oakland crime reports, 2010 (left), Oakland crime reports 2013 (right) (figure was created by the author) [23].*

#### **Figure 9.**

*Oakland, industrial buildings, lofts/luxury residence, businesses related to artists, yoga and fitness studios, fashionable cafes (figure was created by the author) [23–25].*

**Figure 10.**

*Oakland, industrial buildings, lofts/luxury residence, businesses related to artists overlaid with Amazon Mechanical Turk Delta between infrastructure condition and safety (figure was created by the author) [23, 25].*

#### **Figure 11.**

*Amazon Mechanical Turk Oakland example maps of bicycles, lofts, abandoned buildings, and industrial buildings (figure was created by the author).*

them to the "Amazon Mechanical Turk" system for rating along with a series of questions regarding the content shown in the images. This process takes advantage of human subjectivity when it comes to rating an area based on someone's personal

**165**

**Figure 13.**

*author) [25].*

**Figure 12.**

*the author) [25].*

*Human-Centered Approaches in Urban Analytics and Placemaking*

*Amazon Mechanical Turk neighborhood rating: neighborhood infrastructure evaluation (figure was created by* 

*Amazon Mechanical Turk neighborhood rating: neighborhood safety evaluation (figure was created by the* 

*DOI: http://dx.doi.org/10.5772/intechopen.89675*

*Human-Centered Approaches in Urban Analytics and Placemaking DOI: http://dx.doi.org/10.5772/intechopen.89675*

#### **Figure 12.**

*Sustainability in Urban Planning and Design*

**164**

**Figure 11.**

*buildings (figure was created by the author).*

**Figure 10.**

*Oakland, industrial buildings, lofts/luxury residence, businesses related to artists overlaid with Amazon Mechanical Turk Delta between infrastructure condition and safety (figure was created by the author) [23, 25].*

*Amazon Mechanical Turk Oakland example maps of bicycles, lofts, abandoned buildings, and industrial* 

them to the "Amazon Mechanical Turk" system for rating along with a series of questions regarding the content shown in the images. This process takes advantage of human subjectivity when it comes to rating an area based on someone's personal *Amazon Mechanical Turk neighborhood rating: neighborhood infrastructure evaluation (figure was created by the author) [25].*

#### **Figure 13.**

*Amazon Mechanical Turk neighborhood rating: neighborhood safety evaluation (figure was created by the author) [25].*

infrastructure condition (**Figure 12**), interpretation of safety (**Figure 13**) and affordability (**Figure 14**), qualities that vary significantly even among neighboring blocks; however, the amount or the frequency of variation may have a significant role in the overall research. The second set of questions is related to the collection of detail features, such as the presence of expensive loft housing, abandoned buildings, industrial buildings, trees, fitness studios, contemporary and stylish coffee shops. This process is utilizing the same strategy as the first one, by using Google Street viewpoints in order for the participants to identify the presence of any of the feature elements in the content of the images. The identification of these features would be extremely time consuming to collect manually; therefore, this method is proven highly efficient on this aspect. The areas of interest for both sets of questionnaires are Oakland and Emeryville, which were chosen because they are transforming from high concentrated crime areas into urban, entertainment, and commercial attractor points. The questions were submitted to "Amazon Mechanical Turk" as a file in ".json" format and were structured in a way that the answers would be easy to process and to visualize. In particular, the answers to the question would have to be represented either as a numerical scale from 1 to 10, as a binary yes or no option, or

**167**

*Human-Centered Approaches in Urban Analytics and Placemaking*

a multiple choice (tick the box). We avoided completely answers that would require the user to write lengthy texts (**Figure 15**). The received answers were in ".json" format, so they were transformed into ".csv" format as in the previous method.

*Amazon Mechanical Turk submitted questionnaire (left) Amazon Mechanical Turk street rating (right)* 

All data layers were combined and provided the context for a more fine-grained understanding of neighborhood characteristics, conflicts, and relationships that reveal the heterogeneous characteristics of the city [15]. Mapping here is not only addressed as a visualization tool but also as a platform based on which we can make

Moving away from the expert urbanist model, which determines the form and functionality of the built environment based on central rules, we argue that engagement with democratic participation can lead to more sustainable and resilient built environments. "Openreblock" platform is an open-ended approach to social justice that offers users active participation and opportunities to reform their immediate environment (**Figure 16**). By encouraging participatory planning via community mapping by its own citizens, it contributes in improving slum communities and their integration in the broader urban fabric. Some of the immediate benefits are land regularization and security of land ownership, allowance for public services, and connectivity.

As urban planning should be understood as a communicative, pragmatic, social practice, this tool facilitates intercultural dialog and implementation. "Openreblock" enables users to reorganize slum communities that lack significant public infrastructure, such as access to a public street. The idea of the tool is that citizens have the right to affect the design of their local neighborhood and have access to an open-source methodology for doing so. It is a web-based service for an open-source platform that proposes the least disruptive reformation of the existing street network

*DOI: http://dx.doi.org/10.5772/intechopen.89675*

faster and factual assessments [16].

*(figure was created by the author) [25].*

**6.1 Introduction**

**Figure 15.**

**6. Case study 2: urban placemaking through user input**

#### *Human-Centered Approaches in Urban Analytics and Placemaking DOI: http://dx.doi.org/10.5772/intechopen.89675*

#### **Figure 15.**

*Sustainability in Urban Planning and Design*

infrastructure condition (**Figure 12**), interpretation of safety (**Figure 13**) and affordability (**Figure 14**), qualities that vary significantly even among neighboring blocks; however, the amount or the frequency of variation may have a significant role in the overall research. The second set of questions is related to the collection of detail features, such as the presence of expensive loft housing, abandoned buildings, industrial buildings, trees, fitness studios, contemporary and stylish coffee shops. This process is utilizing the same strategy as the first one, by using Google Street viewpoints in order for the participants to identify the presence of any of the feature elements in the content of the images. The identification of these features would be extremely time consuming to collect manually; therefore, this method is proven highly efficient on this aspect. The areas of interest for both sets of questionnaires are Oakland and Emeryville, which were chosen because they are transforming from high concentrated crime areas into urban, entertainment, and commercial attractor points. The questions were submitted to "Amazon Mechanical Turk" as a file in ".json" format and were structured in a way that the answers would be easy to process and to visualize. In particular, the answers to the question would have to be represented either as a numerical scale from 1 to 10, as a binary yes or no option, or

*Amazon Mechanical Turk neighborhood rating: neighborhood affordability evaluation (figure was created by* 

**166**

**Figure 14.**

*the author) [25].*

*Amazon Mechanical Turk submitted questionnaire (left) Amazon Mechanical Turk street rating (right) (figure was created by the author) [25].*

a multiple choice (tick the box). We avoided completely answers that would require the user to write lengthy texts (**Figure 15**). The received answers were in ".json" format, so they were transformed into ".csv" format as in the previous method.

All data layers were combined and provided the context for a more fine-grained understanding of neighborhood characteristics, conflicts, and relationships that reveal the heterogeneous characteristics of the city [15]. Mapping here is not only addressed as a visualization tool but also as a platform based on which we can make faster and factual assessments [16].
