Erik Lehmann


2024

pdf bib
Climate Policy Transformer: Utilizing NLP to track the coherence of Climate Policy Documents in the Context of the Paris Agreement
Prashant Singh | Erik Lehmann | Mark Tyrrell
Proceedings of the 1st Workshop on Natural Language Processing Meets Climate Change (ClimateNLP 2024)

Climate policy implementation is pivotal inglobal efforts to mitigate and adapt to climatechange. In this context, this paper explores theuse of Natural Language Processing (NLP) as atool for policy advisors to efficiently track andassess climate policy and strategies, such asNationally Determined Contributions (NDCs).These documents are essential for monitoringcoherence with the Paris Agreement, yet theiranalysis traditionally demands significant la-bor and time. We demonstrate how to leverageNLP on existing climate policy databases totransform this process by structuring informa-tion extracted from these otherwise unstruc-tured policy documents and opening avenuesfor a more in-depth analysis of national and re-gional policies. Central to our approach is thecreation of a machine-learning (ML) dataset’CPo-CD’, based on data provided by the Inter-national Climate Initiative (IKI) and ClimateWatch (CW). The CPo-CD dataset is utilizedto fine-tune Transformer Models on classify-ing climate targets, actions, policies, and plans,along with their sector, mitigation-adaptation,and greenhouse gas (GHG) components. Wepublish our model and dataset on a HuggingFace repository.

2020

pdf bib
Bilingual Transfer Learning for Online Product Classification
Erik Lehmann | András Simonyi | Lukas Henkel | Jörn Franke
Proceedings of Workshop on Natural Language Processing in E-Commerce

Consumer Price Indices (CPIs) are one of the major statistics produced by Statistical Offices, and of crucial importance to Central Banks. To calculate CPIs, statistical offices collect a large amount of individual prices of goods and services. Nowadays prices of many consumer goods can be obtained online, enabling a much more detailed measurement of inflation rates. One major challenge is to classify the variety of products, from different shops and languages into the given statistical schema consisting of a complex multi-level classification hierarchy - the European Classification of Individual Consumption according to Purpose (ECOICOP) for European countries, since there is no model, mapping or labelled data available. We focus in our analysis on food, beverage and tobacco which account for 74 of the 258 ECOICOP categories and 19 % of the Euro Area inflation basket. In this paper we build a classifier on web scraped, hand-labeled product data from German retailers and test the transfer to French data using cross lingual word embedding. We compare its performance against a classifier trained on the single languages and a classifier with both languages trained jointly. Furthermore, we propose a pipeline to effectively create a data set with balanced labels using transferred predictions and active learning. In addition we test how much data it takes to build a single language classifier from scratch an if there are benefits from multilingual training. Our proposed system reduces the time to complete the task by about two thirds and is already used to support the analysis of inflation.