Sami Diaf

Dr. Sami Diaf

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 January 2022 in European Journal of Political Economy
Verified
Journal Article
Closed access
SDSami Diaf
Sami Diaf
JDJörg Döpke
Jörg Döpke
UFUlrich Fritsche
Ulrich Fritsche
IRIda Rockenbach
Ida Rockenbach

16 November 2022 in Algorithms
Verified
Journal Article
Open access
PDF file
SDSami Diaf
Sami Diaf
UFUlrich Fritsche
Ulrich Fritsche
This paper proposes a new methodology to study sequential corpora by implementing a two-stage algorithm that learns time-based topics with respect to a scale of document positions and introduces the concept of Topic Scaling, which ranks learned topics within the same document scale. The first stage ranks documents using Wordfish, a Poisson-based document-scaling method, to estimate document positions that serve, in the second stage, as a dependent variable to learn relevant topics via a supervised Latent Dirichlet Allocation. This novelty brings two innovations in text mining as it explains document positions, whose scale is a latent variable, and ranks the inferred topics on the document scale to match their occurrences within the corpus and track their evolution. Tested on the U.S. State Of The Union two-party addresses, this inductive approach reveals that each party dominates one end of the learned scale with interchangeable transitions that follow the parties’ term of office, while it shows for the corpus of German economic forecasting reports a shift in the narrative style adopted by economic institutions following the 2008 financial crisis. Besides a demonstrated high accuracy in predicting in-sample document positions from topic scores, this method unfolds further hidden topics that differentiate similar documents by increasing the number of learned topics to expand potential nested hierarchical topic structures. Compared to other popular topic models, Topic Scaling learns topics with respect to document similarities without specifying a time frequency to learn topic evolution, thus capturing broader topic patterns than dynamic topic models and yielding more interpretable outputs than a plain Latent Dirichlet Allocation.


26 June 2024
Verified
Conference Paper
Open access
PDF file
FSFlorian Schütze
Florian Schütze
SDSami Diaf
Sami Diaf
The influence of specific Google Trends search queries measuring various sentiments on economic performance and stock markets has been extensively documented and used for many purposes. This paper examines the predictive power of queries measuring non-specific sentiment on key macroeconomic variables when linked to a comprehensive sentiment dictionary. The analysis shows that non-specific sentiments do not improve the forecasting quality of the US economy as a whole, except for unemployment, which was found to be predictable for all sentiments. Consequently, the authors suggest that economic-related sentiments with carefully selected words should be used in Google Trends search queries to improve predictive performance. However, if a socio-cultural analysis is to be performed, non-specific sentiments would be suggested, as they can be predicted by the real economic time series of unemployment.

26 June 2024
Verified
Conference Paper
Open access
PDF file
FSFlorian Schütze
Florian Schütze
SDSami Diaf
Sami Diaf
Understanding monetary policy has always been of paramount economic and political importance. However, it remains a difficult task, despite transparency efforts and the regular flow of information to the public, which becomes even more complex when communication channels are multilingual. This paper examines the policy narratives of the Swiss National Bank (SNB) in terms of language and policy preferences, using the corpus of speeches delivered by its members over the period 1997-2022. Using a dynamic semantic search strategy based on top2vec, the framework analysis was able to identify interlingual similarities and differences with the help of pre-trained multilingual models. The results show that the SNB's communication strategy is strongly oriented towards the objectives assigned to the central bank, with attention being paid to systemic risks, banking regulation and financial markets, which emerge as second but no less important objectives, closely linked to the international environment, in particular the Eurosystem as a strategic aspect of the stability of the Swiss franc. The results suggest that English is used exclusively to address core central banking issues (monetary policy, inflation and interest rates), while uncertainty concerns seem to be reported more in German or French. The resulting dual semantic space, consisting of embedded words and documents, yielded relevant topics with respect to the size and scope of the corpus. Furthermore, informative indices could be constructed for policy measurement, as a crisis index was found to be consistent with the business cycle fluctuations and technical recessions experienced in Switzerland over the last 25 years.

28 August 2023 in arXiv
Verified
Preprint
Open access
PDF file
SDSami Diaf
Sami Diaf
Document scaling has been a key component in text-as-data applications for social scientists and a major field of interest for political researchers, who aim at uncovering differences between speakers or parties with the help of different probabilistic and non-probabilistic approaches. Yet, most of these techniques are either built upon the agnostically bag-of-word hypothesis or use prior information borrowed from external sources that might embed the results with a significant bias. If the corpus has long been considered as a collection of documents, it can also be seen as a dense network of connected words whose structure could be clustered to differentiate independent groups of words, based on their co-occurrences in documents, known as communities. This paper introduces CommunityFish as an augmented version of Wordfish based on a hierarchical clustering, namely the Louvain algorithm, on the word space to yield communities as semantic and independent n-grams emerging from the corpus and use them as an input to Wordfish method, instead of considering the word space. This strategy emphasizes the interpretability of the results, since communities have a non-overlapping structure, hence a crucial informative power in discriminating parties or speakers, in addition to allowing a faster execution of the Poisson scaling model. Aside from yielding communities, assumed to be subtopic proxies, the application of this technique outperforms the classic Wordfish model by highlighting historical developments in the U.S. State of the Union addresses and was found to replicate the prevailing political stance in Germany when using the corpus of parties' legislative manifestos.