add News: Data all item

app.R

@@ -47,6 +47,19 @@ load_web_db <- function(){
   df_tag_entity <- db_table_load("wp_esi_tag_entity")
   df_investment <- db_table_load("wp_esi_investment")
   
+  # basic for cross
+  df_technology <- db_table_load("wp_esi_technology")
+  df_industry <- db_table_load("wp_esi_industry")
+  df_function <- db_table_load("wp_esi_function")
+  
+  df_media <- db_table_load("wp_esi_media")
+  df_region <- db_table_load("wp_esi_region")
+  df_type <- db_table_load("wp_esi_type")
+  
+  # news tables
+  df_news <- db_table_load("wp_esi_news")
+  df_news_entity <- db_table_load("wp_esi_news_entity")
+  
   # 1.1.4. disconnect
   dbDisconnect(db)
   
@@ -72,6 +85,19 @@ if (exist_cash()){
   df_tag_entity <- read_df("wp_esi_tag_entity")
   df_investment <- read_df("wp_esi_investment")
   
+  # basic for cross
+  df_technology <- read_df("wp_esi_technology")
+  df_industry <- read_df("wp_esi_industry")
+  df_function <- read_df("wp_esi_function")
+  
+  df_media <- read_df("wp_esi_media")
+  df_region <- read_df("wp_esi_region")
+  df_type <- read_df("wp_esi_type")
+  
+  # news tables
+  df_news <- read_df("wp_esi_news")
+  df_news_entity <- read_df("wp_esi_news_entity")
+  
   # packages management
   detach("package:RMySQL", unload=TRUE)
   library(sqldf)
@@ -131,10 +157,12 @@ body <- dashboardBody(
     
     tabItem(tabName = "company_data_all",
             fluidPage(
+              fluidRow(h2("All data about companies")),
               fluidRow(DT::dataTableOutput("company_data_all_table"))
             )),
     tabItem(tabName = "company_data_tags",
             fluidPage(
+              fluidRow(h2("Data about companies by tags")),
               fluidRow(selectizeInput('company_data_tags_tags', 'Tags', choices = df_tag$name, multiple = TRUE)),
               fluidRow(radioButtons("company_data_tags_tags_filter", "Tags filter",
                        choices = list("OR" = 1, "AND" = 2), selected = 1, inline = T)),
@@ -143,7 +171,7 @@ body <- dashboardBody(
             )),
     tabItem(tabName = "company_analysis_countries",
             fluidPage(
-              fluidRow(h2("Distribution of AI companies by countries")),
+              fluidRow(h2("Analysis of companies by countries")),
               fluidRow(p(tags$b("Number of companies"))),
               fluidRow(
                 column(3, textInput("company_analysis_countries_entities_min", "Min", "100")),
@@ -156,7 +184,7 @@ body <- dashboardBody(
     ),
     tabItem(tabName = "company_analysis_tags",
             fluidPage(
-              fluidRow(h2("Number of companies per tags")),
+              fluidRow(h2("Analysis of companies by tags")),
               fluidRow(selectInput("company_analysis_tags_choice", "Number of tags",
                                    c("> 100", "50:100", "10:50", "5:10", "3:5", "1:3", "1: Max"))),
               fluidRow(plotlyOutput("company_analysis_tags")),
@@ -166,7 +194,7 @@ body <- dashboardBody(
     ),
     tabItem(tabName = "company_analysis_size_tags",
             fluidPage(
-              fluidRow(h2("Average size of company per tags")),
+              fluidRow(h2("Analysis of companies by average size per tags")),
               fluidRow(selectInput("company_analysis_size_tags_choice", "Average size",
                                    c("> 1000", "500 : 1000", "250 : 500", "100 : 250", "50 : 100", "25 : 50", "1 : 25", "1 : Max"))),
               fluidRow(plotlyOutput("company_analysis_size_tags")),
@@ -175,7 +203,7 @@ body <- dashboardBody(
             )),
     tabItem(tabName = "company_analysis_age_tags",
             fluidPage(
-              fluidRow(h2("Average age of company per tags")),
+              fluidRow(h2("Analysis of companies by average age per tags")),
               fluidRow(selectInput("company_analysis_age_tags_choice", "Average age",
                                    c("> 20", "15 : 20", "10 : 15", "7 : 10", "5 : 7", "3 : 5", "1 : 3", "1 : Max"))),
               fluidRow(plotlyOutput("company_analysis_age_tags")),
@@ -187,13 +215,14 @@ body <- dashboardBody(
     
     tabItem(tabName = "invest_data_all",
             fluidPage(
-              fluidRow(h2("All investments")),
+              fluidRow(h2("Data about all investments")),
               fluidRow(DT::dataTableOutput("dt_invest_data_all"))
             )
     ),
     
     tabItem(tabName = "invest_data_tags",
             fluidPage(
+              fluidRow(h2("Data about investments by tags")),
               fluidRow(selectizeInput('invest_data_tags_tags', 'Tags', choices = df_tag$name, multiple = TRUE)),
               fluidRow(radioButtons("invest_data_tags_tags_filter", "Tags filter",
                                     choices = list("OR" = 1, "AND" = 2), selected = 1, inline = T)),
@@ -204,7 +233,7 @@ body <- dashboardBody(
     
     tabItem(tabName = "invest_analysis_tags",
             fluidPage(
-              fluidRow(h2("Investments per tags")),
+              fluidRow(h2("Analysis of investments per tags")),
               fluidRow(uiOutput('invest_analysis_tags_slider')),
               fluidRow(plotlyOutput("invest_analysis_tags")),
               fluidRow(DT::dataTableOutput("dt_invest_analysis_tags")),
@@ -215,6 +244,7 @@ body <- dashboardBody(
     
     tabItem(tabName = "invest_data_tags_date_fund",
             fluidPage(
+              fluidRow(h2("Analysis of investments by tags, dates and fundings")),
               fluidRow(selectizeInput('invest_data_tags_date_fund_tags', 'Tags', choices = df_tag$name, multiple = TRUE)),
               fluidRow(radioButtons("invest_data_tags_date_fund_tags_filter", "Tags filter",
                                     choices = list("OR" = 1, "AND" = 2), selected = 1, inline = T)),
@@ -234,7 +264,8 @@ body <- dashboardBody(
     # 4. News ----
     tabItem(tabName = "news_data_all",
             fluidPage(
-              fluidRow(h2("News"))
+              fluidRow(h2("Data about all news")),
+              fluidRow(DT::dataTableOutput("dt_news_data_all"))
             ))
     
   )
@@ -1500,15 +1531,40 @@ df_company_analysis_age_tags <- reactive({
   # 1. Data frame
   df_news_data_all <- reactive({
     
+    # 1. Feature engineering
+    df_result <- df_news %>% 
+      left_join(df_news_entity, by = c("id" = "news_id")) %>% 
+      left_join(df_entity, by = c("entity_id" = "id")) %>% 
+      left_join(df_media, by = c("media_id" = "id")) %>% 
+      left_join(df_type, by = c("type_id" = "id")) %>% 
+      left_join(df_region, by = c("region_id" = "id")) %>%
+      subset(select = c(id, title, publish_date, url.x,
+                        name.x, name.y, name.x.x, name.y.y)) %>% 
+      rename(url = url.x, entity_name = name.x, media_name = name.y, 
+             type_name = name.x.x, region_name = name.y.y) %>% 
+      mutate(publish_date = as.Date(publish_date))
+    
+    # 2. Get result
+    return(df_result)
+    
+  })
+  
+  # View for DT table
+  df_news_data_all_view <- reactive({
+    df_news_data_all() %>% 
+      mutate(
+        title = paste0(substr(title, 1, 20), "..."),
+        url = paste0(substr(url, 1, 20), "...")
+      ) %>% 
+      subset(select = c(id,title, publish_date, url, entity_name,
+                        media_name, type_name, region_name))
   })
   
   # 1. Table
   output$dt_news_data_all <- DT::renderDataTable(
-    DT::datatable(df_news_data_all(), options = list(pageLength = 10, searching = TRUE), rownames= FALSE)
+    DT::datatable(df_news_data_all_view(), options = list(pageLength = 10, searching = TRUE), rownames= FALSE)
   )
   
 }
 
-
-
 shinyApp(ui, server)
\ No newline at end of file

research_news.R

@@ -2,7 +2,6 @@
 # "News" tables analysis
 #
 
-# "wp_esi_news"
 # "wp_esi_news_entity"
 # "wp_esi_news_radar"
 # "wp_esi_news_function"
@@ -49,4 +48,28 @@ df_function <- db_load("wp_esi_function")
 
 # news tables
 df_news <- db_load("wp_esi_news")
+df_news_entity <- db_load("wp_esi_news_entity")
 
+# other tables
+df_media <- db_load("wp_esi_media")
+df_region <- db_load("wp_esi_region")
+df_type <- db_load("wp_esi_type")
+
+# Feature engineering
+df_news_view <- df_news %>% 
+  left_join(df_news_entity, by = c("id" = "news_id")) %>% 
+  left_join(df_entity, by = c("entity_id" = "id")) %>% 
+  left_join(df_media, by = c("media_id" = "id")) %>% 
+  left_join(df_type, by = c("type_id" = "id")) %>% 
+  left_join(df_region, by = c("region_id" = "id")) %>%
+  subset(select = c(id, title, publish_date, url.x,
+                    name.x, name.y, name.x.x, name.y.y)) %>% 
+  rename(url = url.x, entity_name = name.x, media_name = name.y, 
+         type_name = name.x.x, region_name = name.y.y) %>% 
+  mutate(publish_date = as.Date(publish_date))
+
+# All columns in merge data set
+# write.csv(colnames(df_news_view), "c.txt", row.names = F)
+
+
+# df_news_view <- left_join(df_news_view, df_news_entity, by = c("id" = "news_id"))