Pith. Sign in

RECORD

MPST: A Corpus of Movie Plot Synopses with Tags

PaperPDF from arXiv

data files

Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Sudipta Kar, Suraj Maharjan, A. Pastor López-Monroy, Thamar Solorio

Computation and Language · Submitted 22 Feb 2018

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim We expect the corpus will be useful in other tasks where analysis of narratives is relevant.

how it is stated · hedged possibility 0.70 read the record →

Comment (0) Read on arXiv ↗
Cite ▾
arXiv ID1802.07858 (version 2)
Pith ID2WGNDRYK, Pith's permanent record of this paper and its review
CitationA. Pastor L\'opez-Monroy, Sudipta Kar, Suraj Maharjan, Thamar Solorio. "MPST: A Corpus of Movie Plot Synopses with Tags." arXiv:1802.07858 (2018). https://pith.science/paper/2WGNDRYK
Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 6,133 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.11 · human participants 0.25 · controlled comparison 0.02 · randomized assignment 0.01 · states quantitative result 0.94 · compares with prior results 0.05 · mentions code or data availability 0.11.

The abstract, sentence by sentence

arXiv listing abstract v2. Results and the main claim are underlined; .

Social tagging of movies reveals a wide range of heterogeneous information about movies, like the genre, plot structure, soundtracks, metadata, visual and emotional experiences. Such information can be valuable in building automatic systems to create tags for movies. Automatic tagging systems can help recommendation engines to improve the retrieval of similar movies as well as help viewers to know what to expect from a movie in advance. In this paper, we set out to the task of collecting a corpus of movie plot synopses and tags. We describe a methodology that enabled us to build a fine-grained set of around 70 tags exposing heterogeneous characteristics of movie plots and the multi-label associations of these tags with some 14K movie plot synopses. We investigate how these tags correlate with movies and the flow of emotions throughout different types of movies. Finally, we use this corpus to explore the feasibility of inferring tags from plot synopses. We expect the corpus will be useful in other tasks where analysis of narratives is relevant.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 Social tagging of movies reveals a wide range of heterogeneous information about movies, like the genre, plot structure, soundtracks, metadata, visual and emotional experiences. background 1.00 0.21 0.03 asserted 1.00
s1 Such information can be valuable in building automatic systems to create tags for movies. background 0.94 0.06 0.03 hedged possibility 0.67
s2 Automatic tagging systems can help recommendation engines to improve the retrieval of similar movies as well as help viewers to know what to expect from a movie in advance. background 0.98 0.05 0.04 hedged possibility 0.59
s3 In this paper, we set out to the task of collecting a corpus of movie plot synopses and tags. contribution 1.00 0.86 0.03 asserted 0.89
s4 We describe a methodology that enabled us to build a fine-grained set of around 70 tags exposing heterogeneous characteristics of movie plots and the multi-label associations of these tags with some 14K movie plot synopses. method 0.56 0.97 0.10 asserted 1.00
s5 We investigate how these tags correlate with movies and the flow of emotions throughout different types of movies. contribution 0.90 0.91 0.05 asserted 0.85
s6 Finally, we use this corpus to explore the feasibility of inferring tags from plot synopses. contribution 0.83 0.95 0.08 asserted 0.83
s7 We expect the corpus will be useful in other tasks where analysis of narratives is relevant. future work 0.89 0.15 0.14 hedged possibility 0.70

What kind of paper the classifier reports this is: dataset or benchmark 1.00 · evidence from observation or experiment 0.55.

What was examined, and what was not

sources read
  • arXiv listing abstract v2
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper
Original abstract

Social tagging of movies reveals a wide range of heterogeneous information about movies, like the genre, plot structure, soundtracks, metadata, visual and emotional experiences. Such information can be valuable in building automatic systems to create tags for movies. Automatic tagging systems can help recommendation engines to improve the retrieval of similar movies as well as help viewers to know what to expect from a movie in advance. In this paper, we set out to the task of collecting a corpus of movie plot synopses and tags. We describe a methodology that enabled us to build a fine-grained set of around 70 tags exposing heterogeneous characteristics of movie plots and the multi-label associations of these tags with some 14K movie plot synopses. We investigate how these tags correlate with movies and the flow of emotions throughout different types of movies. Finally, we use this corpus to explore the feasibility of inferring tags from plot synopses. We expect the corpus will be useful in other tasks where analysis of narratives is relevant.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

1 arXiv paper cite this in 1 sentence

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.