Paper: arXiv 2508.15825

Authors: Chenghao Liu, Aniket Mahanti, Ranesh Naha, Guanghao Wang, Erwann Sbai

Abstract

As cryptocurrencies gain popularity, the digital asset marketplace becomes increasingly significant. Understanding social media signals offers valuable insights into investor sentiment and market dynamics. Prior research has predominantly focused on text-based platforms such as Twitter. However, video content remains underexplored, despite potentially containing richer emotional and contextual sentiment that is not fully captured by text alone. In this study, we present a multimodal analysis comparing TikTok and Twitter sentiment, using large language models to extract insights from both video and text data. We investigate the dynamic dependencies and spillover effects between social media sentiment and cryptocurrency market indicators. Our results reveal that TikTok’s video-based sentiment significantly influences speculative assets and short-term market trends, while Twitter’s text-based sentiment aligns more closely with long-term dynamics. Notably, the integration of cross-platform sentiment signals improves forecasting accuracy by up to 20%.

Complexity vs Empirical Score

  • Math Complexity: 4.0/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper employs statistical methods like spillover analysis and regression models with LLMs for sentiment extraction, but lacks deep mathematical derivations or novel theoretical frameworks. Empirically, it is strong with a defined methodology, specific dataset (TikTok/Twitter), and quantitative results (20% accuracy improvement, 35% Dogecoin prediction gain), indicating backtest-ready implementation.

Research Flowchart

  flowchart TD
  Goal["Research Goal:<br>Analyze TikTok & Twitter sentiment<br>impact on crypto market dynamics?"]
  Data["Data Inputs:<br>TikTok (Video) & Twitter (Text)<br>+ Crypto Market Indicators"]
  Method["Methodology:<br>Multimodal Analysis with Large Language Models<br>(Video & Text Feature Extraction)"]
  Process["Computation:<br>Analyze Sentiment Spillover &<br>Dynamic Dependencies between Platforms & Market"]
  Findings["Key Outcomes:<br>1. TikTok sentiment drives short-term trends<br>2. Twitter sentiment aligns with long-term trends<br>3. Cross-platform integration improves accuracy by 20%"]
  
  Goal --> Data
  Data --> Method
  Method --> Process
  Process --> Findings

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a multimodal framework that analyzes TikTok video‑based sentiment alongside Twitter text‑based sentiment to improve cryptocurrency market forecasting.

Source sentence
We introduce a framework for analyzing TikTok video-based sentiment compared to Twitter text-based sentiment in cryptocurrency sentiment analysis.

Assumes 1

  • All time‑series (price, volume, sentiment) are stationary after first differencing. statistical
    Source sentence
    that all-time series are stationary at the first difference. These preliminary results justify the use of DCC-GARCH-based connectedness models in the study

Data

  • Twitter data · Tweepy (Twitter API) text-news public
    Nov 8, 2021 to Mar 13, 2024 · public Tweets related to cryptocurrency
    Source sentence
    Twitter Data: From November 8, 2021 to March 13, 2024, 519,208 public Tweets related to cryptocurrency were collected as JSON line objects using a Tweepy-based web crawler.
  • TikTok data · TikTok API alternative public
    15,348 video posts from top 10 cryptocurrency influencers
    Source sentence
    TikTok Data: A dataset of 15,348 TikTok video posts was collected using a multi step method adapted from Cheng and Li.

Recognised sources: Twitter / X

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • Integrating TikTok and Twitter sentiment improves cryptocurrency return and volume forecasts by up to 20%. predictive-accuracy
    Source sentence
    Our findings show combining TikTok and Twitter sentiment enhances crypto return and volume forecasts by up to 20%, with TikTok alone improving short-term Dogecoin predictions by 35%.
  • TikTok video‑based sentiment strongly predicts short‑term speculative assets such as Dogecoin, while Twitter text‑based sentiment aligns better with long‑term dynamics like Bitcoin. predictive-accuracy
    Source sentence
    TikTok sentiment particularly influences short-term speculative trends, while Twitter sentiment aligns better with long-term market dynamics.
  • Dogecoin is identified as the primary transmitter of volume spillovers, whereas stablecoins show limited responsiveness to social‑media sentiment. statistical
    Source sentence
    Dogecoin is identified as the primary transmitter of volume spillovers, whereas stablecoins exhibit limited responsiveness to social media sentiment.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 22 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.