LinguLink

• 

  Home Screen

Inspiration

At Lingulink, we understand the struggles of language learning first-hand. Our team, which includes three English second language learners, recognizes the difficulties that may arise in language learning endeavors and the lack of resources available to individuals in underfunded institutions. We are dedicated to creating a meaningful and impactful language learning experience that provides users with the opportunity to make a difference in the world. By pairing users with ESL students, Lingulink provides users with the unique opportunity to apply their language skills practically and purposefully, while also making a tangible impact on the lives of others through 1-1 mentorship. Through promoting a collaborative and interactive learning experience that can enhance retention and understanding of language concepts, we hope to bridge the gap between language learners of all levels and endow students to reach their full potential.

What it does

Research has shown that people who teach others are more successful at actually learning and put more effort into their learning process. This app capitalizes on this concept by providing users with the opportunity to teach a child while learning a new language, creating a meaningful and impactful experience that promotes language learning and social impact. By focusing on this unique and research-backed approach, our language-learning app aims to disrupt traditional language-learning methods, which often rely on memorization and rote learning. While there are many language-learning apps and services available, few, if any, offer the same level of real-world, practical application that our app provides.

LinguLink is designed to revolutionize how people learn languages by offering a meaningful and impactful experience. By pairing users with underrepresented children who speak the language they are learning, the app creates a unique opportunity for users to apply their language skills practically and purposefully, while also making a tangible impact on the lives of others.

The app uses state-of-the-art language teaching techniques to provide a comprehensive learning experience that covers grammar, syntax, basic words, and other relevant language skills. Using machine learning and artificial intelligence, the app adapts to each user's learning style and progress, providing a tailored learning experience that is both engaging and effective. By incorporating gamification elements, such as challenges and rewards, and customizable learning paths, the app makes language learning more interactive and enjoyable.

However, what truly sets this app apart is the opportunity for users to apply what they have learned through 1-1 mentorship sessions with underrepresented children. Through these sessions, users can help bridge the gap between language learners of all levels and endow students to reach their full potential. Say you want to learn Spanish. After taking an assessment test, you are directed to a personalized lesson catered towards areas that you need to work on. Upon completion of all the modules, you have the option to take a Spanish proficiency test and once you pass, you will be paired with an ESL student from a lower-income community, who is learning English. You can then use your newfound knowledge to help mentor them and help them reach their learning goals. Not only does this help the student, but it also helps build your own confidence and helps make language learning more meaningful.

How we built it

For our backend, we utilized flask for our server, and firebase’s database to store questions to give out to the user. We used firebase because it was extremely easy to add or remove questions relatively quickly and make changes to it. For front-end code used Html/CSS and Javascript to build the design of the website.

Challenges we ran into

One of the challenges we ran into was determining the appropriate technology stack for the website. Integrating the different technologies we were using to build the website, such as front-end frameworks, back-end languages, and hosting services. This required us to learn new technologies and troubleshoot issues that arose during the development process.

Another challenge we faced was determining the architecture and design of the website. This includes deciding on the layout, user interface, and user experience. Here, creating a functional and visually appealing website that is easy to navigate can be difficult, since we had very limited experience in UX Design and using apps for prototyping such as Figma. Likewise, for our MVP (minimum viable product), we had to make some tough decisions on what features to prioritize and what to leave out due to time constraints. This required careful planning and communication within the team to ensure we were on the same page and making the most efficient use of our time and resources.

Accomplishments that we're proud of

We are proud of the fact that we were able to get a working demo complete after all the errors we had been facing. From developing the prototype through Figma to developing a working product, we have come a long way and have learned a lot throughout the process. We are proud of the fact that we came together as a team and worked together to create a product that we are all proud of.

We were also proud of how well our team worked together at communicating and dividing up the shared work between members.

What we learned

During this Hackathon, we discovered the importance of questioning our idea and being open to feedback and suggestions from others. While our idea of a language-learning app that pairs users with underrepresented children was promising, we recognized the need to refine and improve the idea as they continued to develop it. Through discussions with interviewers at YCombinator and brainstorming sessions, we were able to identify potential areas of weakness and ways to address them, resulting in a stronger and more well-rounded concept. Furthermore, we learned the value of seeking help and guidance from others, especially in areas where we lacked experience such as UX Design. As we encountered challenges in determining the appropriate technology stack for their website or designing a user-friendly interface, we sought out resources and mentors to assist us and this willingness to learn and ask for help ultimately led to a more successful and polished final product.

Overall, we learned about the importance of remaining open-minded and adaptable, as well as the benefits of seeking feedback and assistance from others. By embracing a growth mindset and a willingness to learn and improve through taking criticism, we were able to overcome challenges and develop a strong and impactful product.

What's next for LinguLink

Virtual language immersion: By using augmented or virtual reality, LinguLink could create virtual environments that mimic real-world situations where learners can practice their language skills, such as ordering food at a restaurant, navigating a city, or engaging in conversations with native speakers. This could make the language learning experience more immersive and engaging. Likewise, AR/VR can immerse learners in the culture of the language they are studying, allowing them to explore famous landmarks, museums, and cultural events. This could help learners gain a deeper appreciation of the culture behind the language.

*Introducing live, in-app language classes: * While LinguLink already offers real-world language practice through its language partner feature, adding live, in-app language classes could provide a more structured and comprehensive learning experience, which could appeal to users who prefer a more traditional learning approach.

Offering specialized language courses: To cater to learners with specific language learning needs, LinguLink could consider offering specialized courses, such as business language, medical language, or legal language.

Ethics

When contemplating the issue, it is worth noting that our team comprised three second-language learners of English (ESL). However, only one of these individuals was fortunate enough to have undergone an ESL (English as a Second Language) program. At the same time, the remaining members experienced significant difficulties in their language-learning endeavors. Though given at hand when sharing their experiences as ESL learners, one commonality highlighted amongst the three of us was the failure of resources, that is, the lack of mentorship given underfunded institutions in which we resided. Likewise, the same member who had undergone the program was now a researcher working in AI Ethics working on autonomy, where this principle upholds an individual's right to make their own decisions about their health and well-being.

With LinguLink, we recognize the importance of upholding ethical standards and prioritizing the safety and well-being of all participants, especially when pairing minors with adult language learners. As such, we prioritize confidentiality and informed consent, which are crucial in respecting users' autonomy and ensuring they control their personal information and decision-making processes. By providing transparency about the program and its requirements, we aim to empower users to make informed decisions about their participation in LinguLink. Our ultimate goal is to create a safe and supportive environment where users can engage in language learning with confidence and trust.

However, predictive systems are being used in ways that raise issues of fairness and self-freedom because they make predictions based on information provided by individuals. They can perpetuate biases and might result in unexpected outcomes that the users cannot control. Contemporary research on the ethics of involving minors in online platforms has revealed the imperative for strict safeguards to preserve the rights and welfare of children, such as informed consent and data privacy (Díaz-Pérez et al., 2020). The use of algorithms and machine learning models in decision-making processes that impact children, including educational assessments or child welfare interventions, can raise ethical concerns regarding transparency, accountability, and fairness in the development and deployment of these systems to avert potential harm or discrimination.

In the context of Lingulink, we acknowledge that our approach utilizing gamification and machine learning algorithms to individualize and stimulate language acquisition for learners has limitations, the most prominent of which is algorithmic bias. For example, it has been demonstrated that AI-based platforms can display gender and race biases, even though they are incorporated into the algorithm without explicit intention (Akgun, Greenhow 2021). For example, many Natural Language Models and translating applications such as Google Translate continue to reinforce societal stereotypes (e.g., gender-based labels in different languages) or misuse language (e.g., translating "she" to "he") (Prates et al., 2019).

Additionally, a study by the American Psychology Association shows that one retains information better when expected to teach it to others, calling it The Protégé Effect (Muis et al., 2016). Students lack the skills to conceptualize and solve complex problems resulting in lower scores than those possessing the skills. With that being said, by providing a platform where users can engage in language learning with a partner, LinguLink aims to promote a collaborative and interactive learning experience that can enhance retention and understanding of language concepts.

Furthermore, LinguLink acknowledges that some language learners may need more skills to conceptualize and solve complex problems, resulting in lower scores than those with these skills. By providing a supportive and inclusive learning environment, LinguLink aims to address this disparity and provide opportunities for mentorship and guidance to help students develop these skills. **Addressing these ethical implications.* We understand the ethical implications of emphasizing autonomy and providing users with the right to decide about their personal information and decision-making processes. To address the ethical implications of the system, we are committed to the following:

1. Providing transparency in the program and its requirements empowers users to make informed decisions about their participation. Inspired by Intel's Corporate Responsibility Report, we aim to be transparent about our ethical principles and how they are integrated into our platform's design and operation.
2. Prioritizing confidentiality and informed consent to ensure users are in control of their personal information. We will fully adhere to the General Data Protection Regulation law and only maintain and store necessary personal data.
3. Ensuring fairness and non-discrimination in developing and deploying our algorithms and machine learning models. To mitigate this risk, LinguLink will take a multi-faceted approach. First, we will regularly review and monitor the performance of our algorithms and machine learning models to detect any biases or discriminatory outcomes. This will involve testing the accuracy of the models across different demographic groups and investigating any disparities in performance. Second, we will utilize diverse datasets to train our algorithms and machine learning models. We can reduce the risk of reinforcing or amplifying societal biases by incorporating various data. We will also strive to ensure that our training data is inclusive and representative of the global population to minimize the risk of inadvertently creating a biased model. Third, we will incorporate interpretability and explainability into our models, allowing us to understand how they arrive at their predictions and identify potential sources of bias (Lipton, 2018).

Works Cited

Akgun, and Greenhow. "Artificial Intelligence in Education: Addressing Ethical Challenges in K-12 Settings." AI and Ethics, vol. 2, no. 3, Sept. 2021, pp. 431–40, doi:10.1007/s43681-021-00096-7. APA PsycNet. https://psycnet.apa.org/fulltext/2015-38251-001.html. Accessed 18 Feb. 2023. Díaz-Pérez, et al. "Moral Structuring of Children during the Process of Obtaining Informed Consent in Clinical and Research Settings." BMC Medical Ethics, vol. 21, no. 1, Nov. 2020, pp. 1–10, doi:10.1186/s12910-020-00540-z. Milanesi, Carolina. "Intel Underlines Transparency And Accountability By Sharing Diversity And Inclusion Raw Data." Forbes, 16 May 2022, https://www.forbes.com/sites/carolinamilanesi/2022/05/16/intel-underlines-transparency-and-accountability-by-sharing-diversity-and-inclusion-raw-data/?sh=8d1a78440eb2. Prates et al. "Assessing Gender Bias in Machine Translation: A Case Study with Google Translate." Neural Computing and Applications, vol. 32, no. 10, Mar. 2019, pp. 6363–81, doi:10.1007/s00521-019-04144-6. The Protégé Effect: How You Can Learn by Teaching Others – Effectiviology. https://effectiviology.com/protege-effect-learn-by-teaching/. Accessed 18 Feb. 2023. ACM Digital Library, https://dl.acm.org/doi/pdf/10.1145/3236386.3241340. Accessed 19 Feb. 2023.

Built With

• css
• html
• javascript