The inroads team is looking for a consultant experienced in developing learning algorithms and tools to be used by movement activists and researchers. This person will work collaboratively with the inroads team to build-out the categorization and suggestion functions of the inroads Sentiment Gatherer and Platform for Real-time Sentiment Visualization. The project will include a. the assessment of the current linguistic and sentiment analysis used by the tool; b. developing the algorithms or other technologies to allow users to add to and teach the Sentiment Gatherer; c. contributing to the planned interaction of users with the dictionary, taxonomy, and tool’s learning mechanisms; and d. participating in testing, design, and development review to ensure the user experience and learning elements of the tool go hand in hand.

Required experience or competencies:

• Significant practical experience in machine learning and natural language processing
• Understanding of machine learning techniques and pitfalls.
• Experience with real time data analysis and visualization.
• Experience with Natural Language Processing for classification/categorization
• Available for 1-hour weekly progress and project meeting.
• Iterative design and development process experience and/or comfort.
• Flexible about delivery mechanisms.
• Interested in joining a feminist, communicative, interdependent, and user-design informed team.
• Experience making things that get used.
• Can successfully communicate conceptual ideas and design rationale, both verbally and visually.
• Understands development/IA/UX principles.
• Written and spoken fluency in English for design meeting, process and for natural language processing work.
• Available from principle work between 30 October 2019 and 1 February 2020 (with additional check ins and meetings running through June 2020).

Preferred experience or competencies:

• Experience designing for diverse, global user base.
• Experience with sentiment mapping.
• Experience with Sexual and Reproductive Health and Rights and/or Reproductive Justice movements.
• Experience with art exhibitions, mixed media/visual art.
• Able to meet during some part of US Eastern or Pacific Time 9-5pm.

SCOPE OF WORK

The International Network for the Reduction of Abortion Discrimination and Stigma (inroads) created the “Sentiment Gatherer” to allow individuals to share their personal sentiments about abortion anonymously while adding to a collective, public representation of the group sentiments and associations with abortion. The goal of the project is to use machine learning principles to enable the broader use and deeper real-time analysis of data by users of the inroads Sentiment Gatherer.

The tool was piloted in March 2018 when inroads hosted the 2nd inroads Global Member Gathering in Zagreb, Croatia. Subsequently the tool has been used in a variety of online, local workshops, and global meeting settings throughout Africa, Latin America, Europe, Asia, and North America. The existing dictionary was developed through analysis of community spaces and manually updated based on user interaction. The taxonomy was developed in analysis of community spaces and needs to be applied to the dictionary manually as well. While the manual work was sustainable for small gatherings and a full staff, the desire of inroads members and partners to use the tool themselves and in larger settings has led us to an exciting new phase of development. We are pleased to be working with a large health organization to build the Sentiment Gatherer and Visualizer out into a smarter technology with fewer administrative costs to be used in healthcare, policy, and advocacy training, research, and debate.

Consultant will work as the Natural Language Processing Engineer, contributing to the iterative redesign and development of the inroads Sentiment Gatherer into a self-contained, learning technology.

In collaboration with inroads design and development team, consultant will:

• Attend a meeting team kick off with core planning team and consultant staff to agree on roles, responsibilities, process.
• Participate in weekly check-in calls and roughly bi-weekly worksessions with inroads team to be scheduled for participation from Consultant and inroads Team.
• Deliver all elements to inroads project team and aid in presenting deliverables to the client when needed.
• Review implemented interaction design prior to testing and launch.
• Be available for support or fixes during final testing.
• Participate in follow up activities, including a one-hour meeting debrief with core planning team to capture lessons learned from the platform development, training sessions, and other activities as identified by core planning team for up to one ½ day.

DELIVERABLES

• Assessment of current dictionary, taxonomy, and interface
• NLP Algorithm for categorizing new words and recategorizing and visualizing words in real time
• Testing plan
• Interface and development input

TIMEFRAME

Expected period of performance 30 October 2019 through 15 June 2020, with bulk of the work taking place 30 October 2019- 1 February 2020 (brief participation in final testing and design/development input to continue into mid 2020.

BUDGET

For the above work and any further duties and any additional tasks as agreed between Consultant and project team, the project budget is in the range between $9,500 and $15,500 (USD).

Though inroads does not expect any travel costs for Consultant, the project team will cover reasonable expenses directly associated with the project, to be discussed and approved by project lead prior to expenditures.

EXPRESSING INTEREST

Send the following in an email with “Machine Learning and NLP: Sentiment Gatherer” in the subject line to info@makeinroads.org  by 30 October no later than 1pm EST/ 10am PST/ 6pm GMT.

• Statement expressing your interest and relevant experience
• A project plan or approach to the project scope and goals
• CV and relevant design portfolio or link to portfolio with statement contextualizing the most relevant work to this project
• Day and/or hourly rate
• Availability from 30 October 2019-June 2020

Questions can also be sent to the above email address with a 48 hour turn around time for answers.