Weight Prediction of a Beehive Using Bi-LSTM Network

María Celeste Salas, Hernando González, Hernán González, Carlos Arizmendi, Alhim Vera

Research output: Book / Book Chapter / ReportResearch Bookspeer-review

Abstract

Predicting the future weight of an artificial beehive is fundamental to determine the status and production of an artificial bee beehive, the more weight the beehive has at harvest times the more productive it will be, whether the weight was increased by honey, propolis, royal jelly or brood. This paper presents a bidirectional algorithm (Bi-LSTM) using different configurations and activation functions to obtain different results in order to determine the most accurate prediction for future beehive weight. The models were implemented on a database of an artificial beehive obtained from Kaggle.com, whose location of the beehive is in Würzburg, Germany; the data were taken for the whole year 2017 and the variables obtained from the database are humidity and temperature inside the beehive and beehive weight.

Original languageEnglish
Title of host publicationInformation Technology and Systems - ICITS 2023
EditorsÁlvaro Rocha, Carlos Ferrás, Waldo Ibarra
PublisherSpringer Science and Business Media Deutschland GmbH
Pages285-295
Number of pages11
ISBN (Print)9783031332579
DOIs
StatePublished - 2023
EventInternational Conference on Information Technology and Systems, ICITS 2023 - Cusco, Peru
Duration: 24 Apr 202326 Apr 2023

Publication series

NameLecture Notes in Networks and Systems
Volume691 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

ConferenceInternational Conference on Information Technology and Systems, ICITS 2023
Country/TerritoryPeru
CityCusco
Period24/04/2326/04/23

Keywords

  • Beehive
  • Bi-LSTM network
  • TensorFlow
  • Weight prediction

Fingerprint

Dive into the research topics of 'Weight Prediction of a Beehive Using Bi-LSTM Network'. Together they form a unique fingerprint.

Cite this