Are Topics Interesting or Not? An LDA-based Topic-graph Probabilistic Model for Web Search Personalization

In this article, we propose a Latent Dirichlet Allocation– (LDA) based topic-graph probabilistic personalization model for Web search. This model represents a user graph in a latent topic graph and simultaneously estimates the probabilities that the user is interested in the topics, as well as the probabilities that the user is not interested in the topics. For a given query issued by the user, the webpages that have higher relevancy to the interested topics are promoted, and the webpages more relevant to the non-interesting topics are penalized. In particular, we simulate a user’s search intent by building two profiles: A positive user profile for the probabilities of the user is interested in the topics and a corresponding negative user profile for the probabilities of being not interested in the the topics. The profiles are estimated based on the user’s search logs. A clicked webpage is assumed to include interesting topics. A skipped (viewed but not clicked) webpage is assumed to cover some non-interesting topics to the user. Such estimations are performed in the latent topic space generated by LDA. Moreover, a new approach is proposed to estimate the correlation between a given query and the user’s search history so as to determine how much personalization should be considered for the query. We compare our proposed models with several strong baselines including state-of-the-art personalization approaches. Experiments conducted on a large-scale real user search log collection illustrate the effectiveness of the proposed models.

Graph Neural Collaborative Topic Model for Citation Recommendation

Due to the overload of published scientific articles, citation recommendation has long been a critical research problem for automatically recommending the most relevant citations of given articles. Relational topic models (RTMs) have shown promise on citation prediction via joint modeling of document contents and citations. However, existing RTMs can only capture pairwise or direct (first-order) citation relationships among documents. The indirect (high-order) citation links have been explored in graph neural network–based methods, but these methods suffer from the well-known explainability problem. In this article, we propose a model called Graph Neural Collaborative Topic Model that takes advantage of both relational topic models and graph neural networks to capture high-order citation relationships and to have higher explainability due to the latent topic semantic structure. Experiments on three real-world citation datasets show that our model outperforms several competitive baseline methods on citation recommendation. In addition, we show that our approach can learn better topics than the existing approaches. The recommendation results can be well explained by the underlying topics.

CATS: Customizable Abstractive Topic-based Summarization

Neural sequence-to-sequence models are the state-of-the-art approach used in abstractive summarization of textual documents, useful for producing condensed versions of source text narratives without being restricted to using only words from the original text. Despite the advances in abstractive summarization, custom generation of summaries (e.g., towards a user’s preference) remains unexplored. In this article, we present CATS, an abstractive neural summarization model that summarizes content in a sequence-to-sequence fashion while also introducing a new mechanism to control the underlying latent topic distribution of the produced summaries. We empirically illustrate the efficacy of our model in producing customized summaries and present findings that facilitate the design of such systems. We use the well-known CNN/DailyMail dataset to evaluate our model. Furthermore, we present a transfer-learning method and demonstrate the effectiveness of our approach in a low resource setting, i.e., abstractive summarization of meetings minutes, where combining the main available meetings’ transcripts datasets, AMI and International Computer Science Institute(ICSI) , results in merely a few hundred training documents.

Evaluasi Topik Tersembunyi Berdasarkan Aspect Extraction menggunakan Pengembangan Latent Dirichlet Allocation

Recently, Sentiment Analysis is used for expression detection of products or services. Sentiment Analysis is one category type with a level of aspect focused on extracting product aspects. One of the common methods used for aspect extraction is Latent Dirichlet Allocation (LDA) using random topic identification, but this method has not been able to find an acceptable topic with some aspects having been found. Undeterminable topics are referred to as the hidden topics. This study purpose is to evaluate and compare the suitability of identifying hidden topics between human and computer evaluation. The study is also focused on aspect extraction using a variety of LDA innovations. The data used in this study used case studies on e-Commerce. Data were processed using feature selection and grouped using LDA development. Then the data results are processed using Latent Topic Identification based on subjective and objective evaluations. The identification of hidden topic results was evaluated using several semantic and lexicon tests. The evaluation results indicate the comparison of two hidden topic identification assessment values is quite relevant with the average difference in value reaching 6%. As a result, computer calculations assist humans in determining topics if each topic has a low coherence value.

Analysis and Visualization Latent Topic on COVID-19 Vaccine Tweet use two-stage topic modeling (Preprint)

BACKGROUND The World Health Organization (WHO) declared COVID-19 as a global pandemic on January 30, 2020. However, the pandemic has not been over yet. Furthermore, in the first quartal of 2021, some countries face the third wave of the pandemic. During the difficult time, the development of the vaccines for COVID-19 accelerates rapidly. Understanding the public perception of the COVID-19 Vaccine according to the data collected from social media can widen the perspective on the state of the global pandemic OBJECTIVE This study explores and analyzes the latent topic on COVID-19 Vaccine Tweet posted by individuals from various countries by using two-stage topic modeling. METHODS A two-stage analysis in topic modeling was proposed to investigating people’s reactions in five countries. The first stage is Latent Dirichlet Allocation that produces the latent topics with the corresponding term distributions that facilitate the investigators to understand the main issues or opinions. The second stage then performs agglomerative clustering on the latent topics based on Hellinger distance, which merges close topics hierarchically into topic clusters to visualize those topics in either tree or graph views. RESULTS In general, the topic discussion regarding the COVID-19 Vaccine in five countries is similar. Topic themes such as "first vaccine" and & "vaccine effect" dominate the public discussion. The remarkable point is that people in some countries have some topic themes, such as "politician opinion" and " stay home" in Canada, "emergency" in India, and & "blood clots" in the United Kingdom. The analysis also shows the most popular COVID-19 Vaccine, which is gaining more public interest. CONCLUSIONS With LDA and Hierarchical clustering, two-stage topic modeling is powerful for visualizing the latent topics and understanding the public perception regarding the COVID-19 Vaccine.

Mining heterogeneous clinical notes by multi-modal latent topic model

Latent knowledge can be extracted from the electronic notes that are recorded during patient encounters with the health system. Using these clinical notes to decipher a patient’s underlying comorbidites, symptom burdens, and treatment courses is an ongoing challenge. Latent topic model as an efficient Bayesian method can be used to model each patient’s clinical notes as “documents” and the words in the notes as “tokens”. However, standard latent topic models assume that all of the notes follow the same topic distribution, regardless of the type of note or the domain expertise of the author (such as doctors or nurses). We propose a novel application of latent topic modeling, using multi-note topic model (MNTM) to jointly infer distinct topic distributions of notes of different types. We applied our model to clinical notes from the MIMIC-III dataset to infer distinct topic distributions over the physician and nursing note types. Based on manual assessments made by clinicians, we observed a significant improvement in topic interpretability using MNTM modeling over the baseline single-note topic models that ignore the note types. Moreover, our MNTM model led to a significantly higher prediction accuracy for prolonged mechanical ventilation and mortality using only the first 48 hours of patient data. By correlating the patients’ topic mixture with hospital mortality and prolonged mechanical ventilation, we identified several diagnostic topics that are associated with poor outcomes. Because of its elegant and intuitive formation, we envision a broad application of our approach in mining multi-modality text-based healthcare information that goes beyond clinical notes. Code available at https://github.com/li-lab-mcgill/heterogeneous_ehr.