Well, Why we need research papers?
Research paper gives theoretical and practical implementation also deep knowledge about a certain topic.
Every academic student of any field, as well as university professors wishes to publish research papers. Inorder to get prestige, establishing credentials to make a way for future career or to help one get a future grant proposal and so on.
|The Research Cycle
To write a research paper about a certain topic, one has to get the whole background knowledge about the research topic. Need to start web search by crawling several related research papers from search engines, search tools, etc. After doing analyse, calculate, experiment, observing several web pages, we feel whatever research done till now is not relevant. Then again we start the hunt for more several web pages which consume a lot of time just to acquire the relevant information about any scientific research abstract.This makes the researchers frustrate and loss of good quality time in just surfing the web for hours.
What is the Challenge?
So the challenge here is how we can get the research related, relevant information links in just a second? Many researchers, authors face time related issues just to search the relevant Research papers related to their formulated topic in the web of information complexity.
Here’s the proposed solution by iris.ai team:
They launched a tool named as iris.ai which is a scientific research assistant using Artificial Intelligence technique. This Ai tool help the anyone that wants to find related relevant papers for an original research question.
Iris.ai Team :
The concept for Iris.ai was first established three years ago at NASA Ames Research Centre. The team was taking part in a summer programme run by Singularity University (SU) when they were set the task of creating a concept that would positively affect the lives of a billion people. This exercise got the team thinking about the current state of scientific research, and more specifically about the restrictions created by paywalls, and the inability of human intelligence alone to process the three thousand or so research papers that are published around the world every single day.
In May 2016 Iris.ai team launched their first version of AI training tool to help Iris.ai learn via Ted Talks.To increase the accuracy level of the Iris.ai, they needed large datasets which can continuously train the model. Since then they have asked their users cum AI Trainers to join the community and to make a big contribution by participating in the learning experiment. The crowd-training soon became the backbone of Iris.ai by training thousands of texts.This crowd-training of the model doesn’t just improve the accuracy of the algorithm by several percentages, but it also verifies and assesses the quality of the model.
Maria Ritola, co-founder of Iris.ai, talked to The Saint about the software, and the targets that the team are currently working towards.
“Our motivation at Iris.ai is really to build tools that help researchers to leverage the existing scientific knowledge in the research process much more efficiently,” she explained.
Working of Iris.ai:
Are you interested in knowing how the tool- Iris.ai works?
Well, at the moment company does that through specific tools incorporated within the Iris.ai framework.
The first tool is the exploration tool, which allows anyone to input a research paper full text description or any ted talk links. Then the tool bypasses the related queries,abstract, description, and citations. The system gets clear understanding of all inputs thus and exports the related relevant research papers reading list.
The second tool, focuses on narrowing down the reading list to more accurate and relevant papers to select one. So the focus tool, semi automates the literature review process. It goes through the thousands of piece of papers by asking to mark and map the concepts for inclusive and exclusive.Thus, the tool gets trained to find the relevant piece of paper. Such semi-autonomous processes actually help to minimise the searching hours and to process much faster.Because then they can actually focus on reading, analysing, figuring out hypothesis, combining knowledge rather than searching manually taking vast amount of time through thousands of papers available on the web.
To make the Software more smarter in his work, Iris.ai team uses two approaches.The first approach is supervised machine Learning which involves 10,000 people training the software model directly, taking abstract as input and understanding what the trainer think is the most relevant part of these articles. The second approach is unsupervised machine Learning for training Iris.ai where the software model learns directly from research article without the trainer.
How to? For trainers:
Any Users or AI trainers can train the AI system through online by registering as a Iris.ai Trainer.
Here’s the youtube link which guides about how to try Iris.ai on your favourite research paper. To Sign Up and contribute your research abstracts by being an AI trainer Click here –>Iris.ai
Their next goal is to gather and inject a trained dataset of 5000 paper abstracts to the algorithm. And to have 10% improvement in the connection of neural nets with those inputs.
Another development that the team are working towards is the creation of a software that’s capable of forming and testing its own hypothesis – something that would make the Iris.ai software a researcher in its own right.
Also, the team hope to expand their collaboration with universities. So far the Iris.ai team is collaborating with a number of Scandinavian universities and other continental European research partners, including Aalto University, the University of Helsinki and Chalmers University.