Development of a Model for Predicting the Optimal Value for Selected Dataset Using a Web Application

2022

Закрытый доступ Метаданные (CSL-JSON)
DOI 10.1109/sist54437.2022.9945803
Достоверность данных 70 средняя
Как считается эта оценка

Складывается из проверяемых признаков самой записи: есть ли у статьи DOI, сколько независимых источников подтвердили её данные, заполнены ли обязательные поля и насколько свежо последнее подтверждение. Снижают оценку расхождения между источниками и отсутствие идентификатора. Значение от 80 означает, что запись опубликована без дополнительной проверки; от 60 до 79 — что она опубликована с пониженной достоверностью и ручная проверка желательна. Запись, выверенную редактором вручную, пересчёт не опускает ниже 90. Это оценка наших данных о статье, а не оценка самой статьи.

Аннотация

Data analysis tools, data visualization, and the ability to predict the optimal value or obtain the desired result are becoming more and more necessary for people. This need arises both for professional analysts, scientists, specialists who work with data daily, and for people who do not have such deep knowledge to work with data. In this regard, we studied and analyzed works that helped people in solving emerging problems when working with data. Thus, we have identified the need to create a platform that could provide the necessary tools and algorithms to resolve the issues that have arisen. Therefore, we developed a prototype web platform based on the Streamlit library and developed a model to predict the optimal value based on the data received. Building the model required us to study and analyze the appropriate scientific methods. In the developed platform, the user can upload data, visualize them in the way they need, can choose machine learning methods, and receive with our algorithm a forecast to obtain the optimal value. At the end of our work, we tested the constructed model to calculate the accuracy, which showed satisfactory results. As a result, we have developed an optimal value analysis and prediction model, which currently works best for predicting real estate prices. We used this model on the developed web platform, where we use an understandable, user-friendly interface and provide the fundamental tools necessary to work with data.