When was the last time you went out of your way to explore different perspectives on a news story? If you’re like most people, probably never. In fact, odds are that you’re reading news from the same five sources every day. This is by design: social media recommendation algorithms are built to personalize content based on your interests.
In today's world, where most of the news is consumed through social media algorithms, the problem of getting trapped in a filter bubble has become increasingly prevalent. For example, a study conducted in 2020 found that 65% of people reported that the news they read had a significant impact on their political beliefs. Another study showed that up to 80% of people rely on the media for financial decision-making, with news articles about stocks playing a significant role in their investment choices.
These statistics highlight the powerful influence that the media can have on our beliefs and decisions. With a large portion of the population being swayed by the news, it's more important than ever to provide a balanced view of the information, free from political , financial, and any other biases. Duality is working to address these issues by providing users with a broader perspective on the news, breaking down echo chambers and promoting greater understanding.
The Duality extension suggests articles based on a combination of the sentiment analysis of keywords and the overall sentiment of an article as determined by the Google Cloud NLP API. This information is displayed in an easy-to-use extension popup located on the bottom right of the screen, providing users with quick access to articles that offer opposing perspectives on the topic they are reading about.
At Duality, we have developed a comprehensive dashboard platform that provides advanced data analytics on the information collected from our users. With this platform, we aim to deliver a more in-depth understanding of the articles read and recommended through our extension, enabling our users to gain insights and valuable information from the data gathered.
The Duality Score helps users monitor their news consumption and determine if they have a biased reading pattern. The score is based on the frequency of utilising Duality browser extension suggestions and reading opposing articles. It measures the user's news diversity and highlights areas for improvement. With the Duality Score, users can track their progress and make informed decisions to broaden their news perspectives and escape their filter bubble.
Duality also provides a convenient search option for its users to keep track of all the articles they have read and those that were recommended to them. This feature allows users to easily revisit any article they may have come across during their time using the platform. By searching for a specific article, users can quickly access its contents and gain a better understanding of the opposing views that were presented to them. This feature also makes it easy for users to compare different articles and evaluate their own opinions on a given subject.
Duality's word cloud feature is a unique and innovative addition to the platform's dashboard. The word cloud displays the most frequently used keywords from the articles read by the user while using the Duality extension, including the articles recommended by the platform. This feature provides a visual representation of the topics and themes that are most relevant to the user's reading habits.
Duality has an implemented feature that offers a daily recap of news articles based on the user's search history data. These articles are displayed as news cards and have been analyzed for sentiment to ensure that they fall within a neutral range. The daily recap provides a quick and convenient way for users to stay up-to-date on current events, making it an indispensable tool for anyone looking to stay informed in today's fast-paced world.
Duality also includes a functionality that allows users to track the sentiment of specific topics in the articles they visit. With this feature, users can select a specific topic and view a *plot of the average sentiment * of that word across various articles. This provides users with an in-depth understanding of how the sentiment around a topic has evolved over time and helps them track changes in public opinion.
The backend of Duality is built using Flask, a popular Python framework for web development, and MongoDB for data storage. The Google Cloud Natural Language Processing API and BeautifulSoup library are used for sentiment analysis and data scraping respectively. The front-end of the extension is built using React and NextJS for server-side rendering and component-based architecture. Framer Motion is used for animations, and D3.js is used for creating charts and graphs.
Developing Duality project was a challenging experience. One challenge was integrating Google NLP API into the backend, which required a deep understanding of the API and its features. We also faced the challenge of designing a user-friendly interface for the browser extension and website. Another challenge was incorporating the word cloud feature. We needed to find a way to process the vast amount of data generated from user interactions with the extension and present it in a way that was meaningful and easy to understand. This involved developing algorithms to identify and display the most relevant keywords, while also ensuring that the word cloud was visually appealing and engaging for users. Finally, we needed to find a way to effectively visualize the sentiment analysis data, which required a deep understanding of data visualization and the use of D3.js and Framer Motion technologies.
One of our accomplishments in the development of Duality was the successful integration of advanced technologies such as Google NLP API and D3.js. Our sentiment analysis capabilities provide users with a well-rounded understanding of the news and help them identify potential biases. Additionally, the Duality Score has been well received, as it allows users to track and measure the diversity of their news consumption. By tracking the sentiment of their news consumption, users can gain a deeper understanding of their reading habits and take steps towards a more balanced and informed view of the world. In addition to the Duality Score, the extension is also collecting user data on reading habits and bias, as well as other parameters. This rich data allows us to gain valuable insights into the reading behavior of individuals and the media landscape as a whole. This information can then be used to improve the overall user experience, inform media companies and content creators on how to best reach and engage their audience, and even inform public policy decisions related to media literacy and information consumption. By generating this kind of data, Duality is not only helping users expand their views, but also contributing to a better-informed society.
During the development process, we have gained insights into the importance of media literacy and the impact of news consumption on people's beliefs and decision-making. We have also learned the significance of presenting a diverse and balanced view of information to help people break out of their filter bubbles and gain a more nuanced understanding of the world. In addition, we have gained a deeper understanding of the technical aspects of building a browser extension and working with data analysis. We have learned how to gather, process, and visualise data to provide insights into users' reading habits and help them make informed decisions. The user data we have generated will also provide valuable insights for future research and development of the project. Overall, working on the Duality project has been a rewarding experience that has taught us the importance of media literacy and the power of technology to help promote greater understanding and diversity in news consumption.
One future direction for Duality could be to incorporate machine learning and natural language processing algorithms to further enhance its recommendation system. Duality's advanced analytics capabilities can provide valuable insights into reading trends and habits, which can be leveraged by media outlets and publishers to improve their content offerings and target specific audiences. For example, Duality could help publishers understand which topics are resonating with their readers, allowing them to create more targeted content that appeals to their audience's interests. With its innovative approach to news consumption and advanced data analytics capabilities, Duality is well positioned to form long-lasting and mutually beneficial partnerships with companies seeking to better understand their audience and deliver tailored content to meet their needs.