× This Challenge was posted 4 years ago

Challenge view

Back to Project

9: 360° Stakeholderfeedback

Stakeholderanalysis dashboard to mitigate risks in building projects.

⛶  Fullscreen ↓  Download

Development of a tool for the 360° analysis of stakeholder interests and identification of key opinion leaders (KOLs)

WHY

For large projects with high level of investments such as building projects delays and changes during the planning stages are a risk. One of the measurements to mitigate this is to closely manage the different stakeholders to identify hidden requirements and problems as early as possible. IVO Innenentwicklung AG advises and supports municipalities and cities in the transformation of urban spaces. In order for IVO to meet the interests of as many stakeholders as possible, their needs and requirements are assessed by means of personal interviews and online surveys, which serves as a good data basis but does not provide a quick overview on status, issues and stakeholder group.

BUSINESS VALUE

Better understanding of stakeholder and community needs Support in opinion forming with a tailor-made communication strategy by identifying KOLs Reputation building and expectation management through insights gained from stakeholder management Efficient and structured approach to data analysis



WHAT

We are addressing this problem by providing a concise but informative dashboard containing grouping, qualification and prioritization of the stakeholders in order to make more of the available data and provide users with a fast overview by project.

HOW

Aggregation multiple data sources such as data from online surveys (MS Excel), transcripts of personal interviews (MS Word) by NLP sentiment-analysis / part-of-speech-tagging, network analysis, dependency detection and interactive dashboards.

Data sources:

  • Data from online surveys (MS Excel)
  • Transcripts of personal interviews (MS Word)
  • external data sources as required

Aapproaches:

  • NLP: Sentiment-Analysis, Topic Modelling
  • Anomaly detection, dependency detection
  • Interactive Dashboards

Technology:

  • Python, R
  • IBM Watson
  • Google NLP API

Repository Link:

Google Colab

NEXT

  • Presentation to and discussion with IVO Innenentwicklung AG
  • Improve Data Collection and set up proper ETL-process
  • User Testings with Dashboard - Initiate Improvement Feedback Cycles
  • Launch MVP

Readme

Hackathon

This jupyter notebook was created for the "Open Data Hackdays - Shape My City" event: https://opendata.ch/projects/hslu_shape-my-city-2020/

Installation

If downloaded locally, install all required modules & create folder structure.

|--root
    |---Data 
        |---interview_n.docx

Usage

Provide your own API key and url when calling IBM Watson in Cell [15].

Data

Put Interviews as .docx files in folder "Data". The analysed csv data will be saved to the current working directory.

CODE

We have written selfdocumenting code, which you can access directly on Google Colab:

CODE on COLAB

DASHBOARD

The interactive Dashboard can be found under the following link:

Interactive Dashboard


This content is a preview from an external site.
 
Contributed 4 years ago by silvan_leibacher for Shape My City
All attendees, sponsors, partners, volunteers and staff at our hackathon are required to agree with the Hack Code of Conduct. Organisers will enforce this code throughout the event. We expect cooperation from all participants to ensure a safe environment for everybody. For more details on how the event is run, see the Guidelines on our wiki.

Creative Commons LicenceThe contents of this website, unless otherwise stated, are licensed under a Creative Commons Attribution 4.0 International License.