Predictive Models of Malicious Behavior in Human Negotiations
Authors: Zahra Nazari, Jonathan Gratch
IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Here we evaluate the formalism against a large corpus of human face-to-face negotiations. We confirm that the model captures how dishonest human negotiators win while seeming fair, even in unstructured negotiations. ... To test these hypotheses, we obtained a large existing corpus of dyadic face-to-face negotiations. 4.2 Results |
| Researcher Affiliation | Academia | Zahra Nazari and Jonathan Gratch Institute for Creative Tachnologies Univeristy of Southern California, Los Angeles California {zahra, gratch}@ict.usc.edu |
| Pseudocode | No | The paper contains mathematical equations and conceptual diagrams, but no structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating the availability of open-source code for the described methodology. |
| Open Datasets | Yes | To test these hypotheses, we obtained a large existing corpus of dyadic face-to-face negotiations. ... The corpus contains negotiations with both an integrative and distributive structure. All negotiations were previously transcribed and annotated, including annotations of offers and preference statements. Thus, the corpus is ideal for testing our hypotheses. 4.1 Corpus Description Negotiation Task: Dyads performed a simulated negotiation exercise known as Auction Wars (see [De Vault et al. 2015]). |
| Dataset Splits | No | The paper describes the corpus and its use for hypothesis testing, including participant numbers (190 participants, analyzed with 146), but does not explicitly detail specific train/validation/test dataset splits with percentages or sample counts for model reproduction. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used for conducting the analysis or experiments. |
| Software Dependencies | No | The paper mentions 'ELAN [Wittenburg et al. 2006]' for annotations but does not provide specific version numbers for this or any other software used in the analysis. |
| Experiment Setup | Yes | 4.1 Corpus Description Negotiation Task: Dyads performed a simulated negotiation exercise known as Auction Wars (see [De Vault et al. 2015]). ... To elicit realistic behavior, negotiators were motivated with a financial reward. ... Annotations: The corpus was manually annotated. All negotiations were segmented, transcribed and semantically annotated by trained coders using ELAN [Wittenburg et al. 2006]. ... 4.3 Measures Hypothesis H1 requires a measure of winnings and a notion of what preferences are communicated. ... Lies: As a measure of explicit lies, we manually annotated all preference statements for their veracity. ... Opponent model (from words): ... we adapt [Nazari et al. 2015] s issue-sentiment opponent-modeling heuristic. ... Opponent model (from deeds): We adapt [Nazari et al. 2015] s issue-ratio heuristic to infer a negotiator s preferences from the offers (DIVs) they make throughout the negotiation. |