Research on Multi-Source Data-Driven Price Forecasting and Decision Support for Food Security Based on AI Large Language Model
- Beschreibung
In recent years, global food security has attracted increasing attention. Under the combined influence of climate change, international trade frictions, and geopolitical uncertainties, food market prices have become more volatile. In particular, against the backdrop of persistent tensions in regions such as the Middle East, geopolitical conflicts have exerted a significant impact on global commodity markets. Meanwhile, such sudden events further intensify uncertainty expectations among market participants and amplify fluctuations in market sentiment.
In this context, food prices are influenced not only by traditional supply–demand relationships, but also by market sentiment and expectation-driven factors. Conventional forecasting methods based on historical prices or a single economic variable struggle to adequately capture price fluctuations triggered by information shocks and changes in sentiment.
Therefore, it is necessary to incorporate news texts and social media data, apply sentiment analysis techniques to extract public opinion signals, and integrate these with market data to construct multi-source data-driven food price forecasting models. This approach can provide more effective tools for risk early warning and decision support in food markets.
- Projektzeitraum
- Sommersemester 2026
- Bewerbungszeitraum
- 07. bis 20.04.2026
- Durchführung
- semesterbegleitend
- Details zu Projektzeitraum und Durchführung
1.All data collection will be completed by 01.06.2026.
2.Data analysis will be conducted and the abstract drafted 01.07.2026.
3.The conference poster will be prepared and finalized by 10.09.2026.
- Studienfach
- offen für alle Studienfächer
- Betreuende
- Luchun Xu
- Institut
- Institut für Kulturpflanzenwissenschaften (340) (Natural Language Processing in Food Security)
- Sprache
- deutsch/englisch
- Teilnehmendenanzahl
- min. 1, max. 4
- Arbeitsaufwand
-
ca. 180 Stunden pro Teilnehmende:r
| 6
ECTS-Punkte
Arbeitsaufwand (Stunden und ggf. ECTS) sind ungefähre Angaben. Die tatsächlich vergebenen ECTS-Punkte ergeben sich aus der tatsächlich geleisteten Arbeit.
- Für dieses Projekt ist kein Motivationsschreiben des Studierenden erforderlich
- Projektart
- experimentell
- Lernziele
-
Die Teilnehmende lernen in diesem Projekt:
1.Literature reading and analysis
2.Data collection
3.Data analysis and data visualization
4.Results presentation (poster, conference)
- Anmerkungen für Studierende
1.A group discussion is held every two weeks to report on the progress of the experiment.
2.Basic understanding of natural language processing and experience with Python and R.