Design of trustworthy online recommendation agents : explanation facilities and decision strategy support
Bibliographic record
Abstract
Due to advances in Web-based technologies, ample opportunities exist to utilize knowledge-based systems for facilitating online consumer decision-making and for providing recommendation services for consumers. This thesis focuses on online recommendation agents that offer shopping advice based on user-specified needs and preferences. Because of the high risks and uncertainties inherent in online environments, effective recommendation agents need to be trustworthy. By extending interpersonal trust to trust in technological artifacts, consumers' trust in a recommendation agent is defined to include three belief components: competence, benevolence, and integrity. This thesis examines user acceptance of online recommendation agents and trust formation in the agents and it empirically investigates agent features and capabilities that increase the trust in them so that a higher chance of user acceptance can be realized. Two important agent capabilities are tested: 1) explanation facilities and 2) decision strategy support. An integrated Trust-TAM (Technology Acceptance Model) was tested and the results show that trust in agents influences consumers' behavioral intentions. Trust in agents exerts a direct impact on the intentions to adopt recommendation agents as well as an indirect impact via the perceived usefulness of the agents. Written protocols were collected and analyzed to identify the major processes that build and inhibit consumers' trust in recommendation agents. The results highlight the important roles of several processes in cultivating and inhibiting agent trust, such as expectation confirmation, utility assessment, and information sharing. Regarding explanation facilities, this research tests three types of explanations - how explanations, why explanations, and guidance. The results indicate that the use of different types of explanations increases different trusting beliefs: the use of how explanations increases competence and benevolence beliefs; the use of why explanations increases the benevolence belief; and the use of guidance increases the integrity belief. The impact of decision strategy support on consumers' trust and adoption of online recommendation agents was also investigated together with explanation facilities. Three types of recommendation agents with different levels of decision strategy support were compared. Both the benefits and costs of providing a high level of decision strategy support were examined. The results suggest that recommendation agents with decision strategy support capabilities and explanation facilities deliver benefits to users (e.g., more useful and trustworthy) and have a higher chance of being adopted by users, when the use of the agents does not require much additional effort. This research has addressed an important gap that exists in our current understanding of trustworthy online recommendation agents. It also makes a key contribution by empirically testing the effects of explanation facilities and decision strategy support on consumers' trust and acceptance of online recommendation agents.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 teacher head, 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".