Bibliographic record
Abstract
The emergence of social platforms has provided us with a vast amount of raw data on people's daily activities, their social connections, interactions, interests, influence, etc. This creates an opportunity to extract valuable insights, helpful in various applications including but not limited to the identification of the trending topics, community detection, health purposes, and marketing. This thesis designs a platform that analyzes users' interactions (e.g., posts, likes, and shares) and connections (e.g., friendship and followership), and maximizes information spread in the network accordingly. There are numerous applications to this such as advertising, knowledge dissemination, encouraging the adoption of productive behavior in a society, etc. I present a 3-part platform that utilizes users' activities to maximize information spread. It starts by analyzing users' posts (Step 1), and proceeds with users' interactions (Step 2), and connections (Step 3). This information is utilized to locate user communities formed based on topical cascades. Targeting these communities helps to increase the spread of information. In the first step, posts (e.g., tweets) are categorized into partitions representing topics of discussion. Part 2 identifies bursty user groups for each topic of interest. A bursty user group includes a collection of socially connected users who are active and influential with regard to a specific topic. Bursty user groups for a topic are the sources of previous activities on that topic and are of high importance in marketing campaigns. Finally having a limited budget for advertising (i.e., a fixed seed set size), Part 3 determines the best groups among all to target to maximize the spread of information in the network. The analysis is based on the state-of-the-art viral marketing models. All of the proposed algorithms have been extensively evaluated on real-world datasets including Twitter fire-hose, co-authorship, and telecommunication datasets. We present quantitative and qualitative experimental results in each chapter attesting to the practicality of our algorithms.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".