{
  "format": "dataset-manifest/1",
  "slug": "ai-powered-customer-service-chatbot",
  "version": 1,
  "seed": "ai-powered-customer-service-chatbot:1",
  "spec": {
    "format": "dataset/1",
    "slug": "ai-powered-customer-service-chatbot",
    "name": "Customer Interaction Insights",
    "tagline": "Customer interactions including inquiries, resolutions, and feedback from AI-driven chatbot communications.",
    "description": "Customer service centers respond to a multitude of inquiries via AI-driven chatbots that streamline communication. These operations involve managing transitions between automated chat systems and human agents, ensuring efficient tracking of agent statuses, and analyzing interactions that occur across various channels. The business focuses on providing timely resolutions to customer queries and improving user satisfaction through continuous feedback mechanisms, including satisfaction surveys. While the majority of interactions result in satisfactory outcomes, there are instances that necessitate escalation for manual resolution, showcasing the nuanced nature of customer service.\n\nThe primary table, customer_interactions, records detailed information about each engagement. Each row in this table represents an individual interaction between a customer and the service center, capturing data such as the nature of the inquiry, the channel used for communication, the outcome of the interaction, and the identifiers for both the chatbot and the human agent involved. This structure allows for comprehensive analysis of customer behavior and the effectiveness of different resolution methods.\n\nBy examining this dataset, developers can explore various questions: What are the most common customer inquiries? How effective are chatbots in resolving issues without human intervention? What factors contribute to customer satisfaction, and how do different channels impact the resolution process? This data is particularly valuable for practicing skills in data analysis, natural language processing, and customer experience optimization.\n\nIt’s important to note that the data is synthetic, designed to emulate real-world customer service interactions while allowing for diverse exploration and analysis.",
    "domain": "social",
    "shape": "graph-edges",
    "useCases": [
      "sql-practice",
      "teaching"
    ],
    "keywords": [
      "customer service",
      "AI chatbot",
      "customer interaction",
      "agent handoff",
      "user satisfaction",
      "interaction analysis"
    ],
    "locale": "en",
    "license": "CC0-1.0",
    "version": 1,
    "tables": {
      "customer_interactions": {
        "description": "This table captures the dynamic exchanges between customers and chatbots, including agent handoff details and user satisfaction feedback. Each row represents a unique interaction instance, linking users to their inquiries and outcomes.",
        "rows": 500,
        "columns": {
          "id": {
            "kind": "pk",
            "description": "Unique identifier for each interaction."
          },
          "customer_id": {
            "kind": "uuid",
            "description": "Identifier for the customer involved in the interaction."
          },
          "agent_id": {
            "kind": "uuid",
            "nullable": true,
            "description": "Identifier for the human agent handling the inquiry (null if not handed off)."
          },
          "interaction_type": {
            "kind": "enum",
            "values": [
              "chatbot",
              "human"
            ],
            "description": "Type of interaction, indicating whether it was handled by the chatbot or a human agent."
          },
          "query": {
            "kind": "sentence",
            "max": 200,
            "description": "The customer's inquiry or request."
          },
          "satisfaction_rating": {
            "kind": "int",
            "min": 1,
            "max": 5,
            "description": "Rating of customer satisfaction with the interaction, on a scale of 1 to 5."
          },
          "escalation": {
            "kind": "bool",
            "trueRate": 0.15,
            "description": "Indicates whether the interaction required escalation to a human agent."
          },
          "created_at": {
            "kind": "datetime",
            "from": "2022-01-01",
            "to": "2023-12-31",
            "description": "Timestamp of when the interaction took place.",
            "dayProfile": "waking"
          },
          "channel": {
            "kind": "enum",
            "values": [
              "web",
              "mobile",
              "social"
            ],
            "description": "The channel through which the customer interacted."
          }
        }
      }
    },
    "structure": "single-table",
    "traits": [
      "categorical",
      "nullable",
      "timestamped"
    ],
    "analysisQuestions": [
      "What are the most common types of customer queries?",
      "How often do interactions escalate to human agents?",
      "What is the average satisfaction rating for chatbot interactions?",
      "Are there specific channels that yield higher satisfaction ratings?",
      "What percentage of interactions are handled solely by chatbots?"
    ],
    "suggestedQueries": [
      {
        "title": "Count of interactions by type",
        "sql": "SELECT interaction_type, COUNT(*) FROM customer_interactions GROUP BY interaction_type;",
        "dialect": "postgres"
      },
      {
        "title": "Average satisfaction rating",
        "sql": "SELECT AVG(satisfaction_rating) FROM customer_interactions;",
        "dialect": "postgres"
      },
      {
        "title": "Satisfaction ratings by channel",
        "sql": "SELECT channel, AVG(satisfaction_rating) FROM customer_interactions GROUP BY channel;",
        "dialect": "postgres"
      },
      {
        "title": "Interactions that escalated",
        "sql": "SELECT COUNT(*) FROM customer_interactions WHERE escalation = TRUE;",
        "dialect": "postgres"
      },
      {
        "title": "Rate of escalation by interaction type",
        "sql": "SELECT interaction_type, COUNT(*) FILTER (WHERE escalation = TRUE) * 100.0 / COUNT(*) AS escalation_rate FROM customer_interactions GROUP BY interaction_type;",
        "dialect": "postgres"
      }
    ],
    "usageScenarios": [
      {
        "title": "Customer Support Training",
        "body": "Trainers can use this dataset to help new customer support agents understand common customer interactions and the importance of timely escalations. By analyzing the data, agents can learn how to better manage customer queries and improve satisfaction ratings."
      },
      {
        "title": "AI Chatbot Development",
        "body": "Developers can utilize this dataset to train AI chatbots by providing them with realistic customer queries and expected responses. This allows for better preparation of the chatbot to handle various scenarios and improve its ability to assist users effectively."
      },
      {
        "title": "Business Analytics",
        "body": "Business analysts can leverage this dataset to identify trends in customer interactions, pinpoint areas for improvement, and evaluate the effectiveness of their customer service strategies. Insights gained can guide decision-making and enhance overall service quality."
      },
      {
        "title": "Research on Customer Behavior",
        "body": "Researchers studying customer behavior in digital environments can apply this dataset to explore how different factors influence customer satisfaction and interaction outcomes. This can lead to valuable insights into user preferences and expectations in online servicing."
      }
    ]
  },
  "counts": {
    "customer_interactions": 500
  },
  "stats": {
    "customer_interactions": {
      "rows": 500,
      "columns": {
        "id": {
          "nulls": 0,
          "distinct": 500,
          "min": 1,
          "max": 500
        },
        "customer_id": {
          "nulls": 0,
          "distinct": 500
        },
        "agent_id": {
          "nulls": 43,
          "distinct": 457
        },
        "interaction_type": {
          "nulls": 0,
          "distinct": 2,
          "min": "chatbot",
          "max": "human"
        },
        "query": {
          "nulls": 0,
          "distinct": 498
        },
        "satisfaction_rating": {
          "nulls": 0,
          "distinct": 5,
          "min": 1,
          "max": 5
        },
        "escalation": {
          "nulls": 0,
          "distinct": 2,
          "min": "false",
          "max": "true"
        },
        "created_at": {
          "nulls": 0,
          "distinct": 500,
          "min": "2022-01-02 11:59:29",
          "max": "2023-12-29 03:23:27"
        },
        "channel": {
          "nulls": 0,
          "distinct": 3,
          "min": "mobile",
          "max": "web"
        }
      }
    }
  },
  "artifacts": [
    {
      "file": "csv/customer_interactions.csv",
      "format": "csv",
      "table": "customer_interactions",
      "key": "datasets/ai-powered-customer-service-chatbot/v1/csv/customer_interactions.csv",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/csv/customer_interactions.csv",
      "bytes": 86553,
      "sha256": "7a7056d67716ec20d8a907b194850cfaef954ad0c2ed0ecb8640214bccb1b72f",
      "rows": 500,
      "contentType": "text/csv; charset=utf-8"
    },
    {
      "file": "ai-powered-customer-service-chatbot.json",
      "format": "json",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/ai-powered-customer-service-chatbot.json",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/ai-powered-customer-service-chatbot.json",
      "bytes": 191375,
      "sha256": "adfbabb31e439434e4c5e0ea0e393258bed021d3c07c161c8885fb37f5edde4a",
      "rows": null,
      "contentType": "application/json; charset=utf-8"
    },
    {
      "file": "sql/postgres.sql",
      "format": "sql-postgres",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/sql/postgres.sql",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/sql/postgres.sql",
      "bytes": 100668,
      "sha256": "1e2bec2bc1277a90c05f2708d5ee41d8890057665de537877c23f7b377934e70",
      "rows": null,
      "contentType": "application/sql; charset=utf-8"
    },
    {
      "file": "sql/mysql.sql",
      "format": "sql-mysql",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/sql/mysql.sql",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/sql/mysql.sql",
      "bytes": 98621,
      "sha256": "755ded0c721500c7dded1670e540ca34080cd73a39a30e5e26a7c46511be45e7",
      "rows": null,
      "contentType": "application/sql; charset=utf-8"
    },
    {
      "file": "sql/mssql.sql",
      "format": "sql-mssql",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/sql/mssql.sql",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/sql/mssql.sql",
      "bytes": 101686,
      "sha256": "e4ed87f42fd739ee282bada184c59b64a57683cb43e9ef97f93654ad7a481990",
      "rows": null,
      "contentType": "application/sql; charset=utf-8"
    },
    {
      "file": "sql/sqlite.sql",
      "format": "sql-sqlite",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/sql/sqlite.sql",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/sql/sqlite.sql",
      "bytes": 98581,
      "sha256": "31f5ced03adc78d0ad0781ba94b3e532aaae3730ab9f45cdcb8ddb01547f80ef",
      "rows": null,
      "contentType": "application/sql; charset=utf-8"
    },
    {
      "file": "prisma/seed.ts",
      "format": "prisma-seed",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/prisma/seed.ts",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/prisma/seed.ts",
      "bytes": 163621,
      "sha256": "5e74d9d343adae73bd78859d1902356f873c425fdd73354a395426ac58acea04",
      "rows": null,
      "contentType": "application/typescript; charset=utf-8"
    },
    {
      "file": "django/fixtures.json",
      "format": "django-fixture",
      "table": null,
      "key": "datasets/ai-powered-customer-service-chatbot/v1/django/fixtures.json",
      "url": "https://database-sample-website.s3.eu-north-1.amazonaws.com/datasets/ai-powered-customer-service-chatbot/v1/django/fixtures.json",
      "bytes": 219842,
      "sha256": "db7ad15c31a396d47f20b77a0d93c0a8ff22ae65e238b10b62e434fe1602ed8e",
      "rows": null,
      "contentType": "application/json; charset=utf-8"
    }
  ],
  "publishedAt": 1788767154643,
  "updatedAt": 1788767154643
}