The perils and pitfalls of explainable AI: Strategies for explaining algorithmic decision-making