Merge pull request #122 from jupyter-naas/120-content---add-new-columns-post_id-company-and-people-mentionned

feat: content engine - add new columns post id company and people mentionned

Author: FlorentRvl

models/content-engine/__pipeline__.ipynb

@@ -2,13 +2,13 @@
  "cells": [
   {
    "cell_type": "markdown",
-   "id": "f60a7fd6-a35b-4d3e-b123-a441e2e0215a",
+   "id": "56b96b04-bd01-4155-92de-45b6f02f0e6f",
    "metadata": {
     "papermill": {},
     "tags": []
    },
    "source": [
-    "<img width=\"8%\" alt=\"Naas.png\" src=\"https://raw.githubusercontent.com/jupyter-naas/awesome-notebooks/master/.github/assets/logos/Naas.png\" style=\"border-radius: 15%\">"
+    "<img width=\"8%\" alt=\"Content\" src=\"https://naasai-public.s3.eu-west-3.amazonaws.com/abi-demo/content_creation.png\" style=\"border-radius: 15%\">"
    ]
   },
   {
@@ -140,6 +140,7 @@
     "# Engine\n",
     "engine_name = \"content-engine\"\n",
     "assistant_name = \"Content Assistant\"\n",
+    "days_start = -7\n",
     "custom_pipeline_path = os.path.join(naas_data_product.MODELS_PATH, engine_name, \"custom\", \"__pipeline__.ipynb\")"
    ]
   },
@@ -220,19 +221,21 @@
    "source": [
     "extraction = NotebookStep(\n",
     "    name=\"LinkedIn\",\n",
-    "    notebook_path=os.path.join(naas_data_product.MODELS_PATH, engine_name, \"core\", \"domain\", \"LinkedIn_Get_entity_posts_stats.ipynb\"),\n",
+    "    notebook_path=os.path.join(naas_data_product.MODELS_PATH, engine_name, \"core\", \"domain\", \"LinkedIn_Get_entity_posts.ipynb\"),\n",
     "    parameters={\n",
     "        \"li_at\": li_at,\n",
     "        \"JSESSIONID\": JSESSIONID,\n",
     "        \"linkedin_url\": linkedin_url,\n",
+    "        \"days_start\": days_start,\n",
     "        \"output_dir\": engine_dir\n",
     "    }\n",
     ")\n",
     "cleaning = NotebookStep(\n",
     "    name=\"Content\",\n",
-    "    notebook_path=os.path.join(naas_data_product.MODELS_PATH, engine_name, \"core\", \"domain\", \"Google_Sheets_Send_content_db_to_spreadsheet.ipynb\"),\n",
+    "    notebook_path=os.path.join(naas_data_product.MODELS_PATH, engine_name, \"core\", \"domain\", \"Content_Create_Posts_database.ipynb\"),\n",
     "    parameters={\n",
     "        \"input_dir\": engine_dir,\n",
+    "        \"entity_dir\": entity_dir,\n",
     "        \"entity_name\": entity_name,\n",
     "        \"spreadsheet_url\": spreadsheet_url,\n",
     "        \"sheet_name\": sheet_posts,\n",

models/content-engine/__plugin__.ipynb

@@ -215,8 +215,22 @@
     "prompt_data = \"\"\n",
     "post_number = 0\n",
     "views = 0\n",
-    "interactions = 0\n",
-    "\n",
+    "engagements = 0\n",
+    "to_keep = [\n",
+    "    \"URL\",\n",
+    "    \"DATE\",\n",
+    "    \"TIME\",\n",
+    "    \"TEXT\",\n",
+    "    \"VIEWS\",\n",
+    "    \"LIKES\",\n",
+    "    \"COMMENTS\",\n",
+    "    \"SHARES\",\n",
+    "    \"ENGAGEMENTS\",\n",
+    "    \"CONCEPT\",\n",
+    "    \"SENTIMENT\",\n",
+    "    \"TARGET\",\n",
+    "    \"OBJECTIVE\",\n",
+    "]\n",
     "# Load data from pickle\n",
     "df = pd.DataFrame()\n",
     "data = pload(input_dir, input_file)    \n",
@@ -228,15 +242,14 @@
     "# Filter on this week and last week\n",
     "if len(df) > 0:\n",
     "    df = df[(df[\"ENTITY\"] == entity_name) & df[\"SCENARIO\"].isin([TW, LW])]\n",
-    "    df[\"INTERACTIONS\"] = df.LIKES + df.COMMENTS + df.SHARES\n",
-    "    prompt_data = df.to_string()\n",
+    "    prompt_data = df[to_keep].to_string()\n",
     "    post_number = len(df)\n",
     "    views = df.VIEWS.sum()\n",
-    "    interactions = df.INTERACTIONS.sum()\n",
+    "    engagements = df.ENGAGEMENTS.sum()\n",
     "    \n",
     "if views == 0:\n",
-    "    kpi = f\"{interactions}\"\n",
-    "    kpi_label = \"interactions\"\n",
+    "    kpi = f\"{engagements}\"\n",
+    "    kpi_label = \"engagements\"\n",
     "else:\n",
     "    kpi = f\"{views}\"\n",
     "    kpi_label = \"impressions\"\n",
@@ -270,10 +283,10 @@
     "system_prompt = f\"\"\"\n",
     "Act as a Content Assistant who has access to valuable data and insights about the content strategy. \n",
     "Your role is to manage and optimize the content, ensuring it reaches the target audience effectively. \n",
-    "This week '{TW}' the '{entity_name}' published {len(df)} contents and generated and {kpi} {kpi_label}. \n",
+    "The last 2 weeks '{TW}' & '{LW}' the '{entity_name}' published {len(df)} contents and generated and {kpi} {kpi_label}. \n",
     "Here is the list with the post published: {prompt_data}.\n",
     "The first message should be about presenting yourself and analyze '{entity_name}' contents published based on best post/topics by {kpi_label}, by date and time with a maximum of 3 bullet points.\n",
-    "Then, you will display the image inside the markdown of the chat about the evolution of content {kpi_label} over the last 2 weeks: [![Content KPI]({image_url})({image_html})].\n",
+    "Then, you will display the image inside the markdown of the chat about the evolution of content {kpi_label} over the last 2 weeks: ![Content KPI]({image_url}).\n",
     "You must ALWAYS show the image in the first message.\n",
     "Be casual, but professional. Wait for the first answer from the user, and then start with the first high-level analysis. \n",
     "\"\"\""
@@ -340,6 +353,14 @@
     "print(\"--> Create new chat\")\n",
     "Markdown(f\"[Create New Chat](https://naas.ai/chat/use?plugin_url={plugin_url})\")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9adc4d13-42d7-4c82-b7b0-621e1bf1c10e",
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {