Search by tag «Text Analysis» 5 results

  • Passing Exams, Writing Papers, and Describing Art: Why Neural Networks Won’t Replace Scientists, Writers and Journalists

    A new version of the ChatGPT chatbot, belonging to the acclaimed GPT-3 family of language models, was published in late November. The developers made significant updates to the chatbot, adding a plethora of new features – for instance, the neural network is now capable of producing texts in a specific style, coming up with arguments that support its statements, and making references to its sources or specific works of literature. As it turns out, it can even write an essay for the Unified State Exam in the Russian language – such an experiment was conducted by philologists from the Higher School of Economics along with Andrey Getmanov, a student of ITMO’s Big Data and Machine Learning Master’s program. We spoke with Andrey to better understand the capabilities of neural networks, and find out whether they will replace scholars, writers, and journalists in the near future.


  • How to Read and Understand Scientific Articles

    If you’re a scientist, keeping up with the latest trends in your industry is a necessity. Researchers have to constantly read scholarly articles from leading scientific journals to know what’s trending, avoid repetition, and make their contribution to science. Yet the amount of reading may feel overwhelming if you don’t know how to quickly process information and extract the main ideas from texts. To help you with that, ITMO.NEWS talked to Dmitriy Tsarev, a junior researcher at ITMO’s Laboratory of Single-Photon Detectors and Generators, who explained the differences between fiction and science literature and gave us a step-by-step guide on how to read scientific papers efficiently.


  • How To Teach a Computer To Interpret Text

    A team of scientists from ITMO University has created a model for an algorithm that can analyze the meaning of texts. It differs from regular neural networks because it not only takes into account word usage frequency but also introduces a category of subjective text perception. Read on to find out what it means for the future of artificial intelligence and how quantum theory is of help here.


  • Neural Networks Predict Future Bestsellers

    ITMO researchers together with a colleague from the University of Oulu in Finland created an algorithm for predicting the success of new books. The algorithm's claimed efficacy is based on the analysis of the emotional fluctuations of a literary text.


  • ITMO Researchers' Improved Text Recognition Algorithm Helps Analyze Century-Old Russian Texts

    Staff members of ITMO University’s Digital Humanities Research Center and Laboratory of Machine Learning joined their forces to teach an algorithm to analyze archival texts and identify names, titles, and addresses. Existing algorithms that were developed using contemporary texts often fail to process books and articles from the past, especially ones that are over a century old. ITMO.NEWS got in touch with the researchers to learn more about this achievement and the St. Retrospect project, a map of the city’s historically significant sites for which this research was conducted.