ai人工智能_人工智能的6种最佳启动选择

2023-11-12

ai人工智能

意见 (Opinion)

Artificial Intelligence is the fastest-growing field in the present-day. According to fortune, the statistics say that the hirings for AI specialists have grown by 74% over the last 4 years. Artificial Intelligence is regarded as the “Hottest” job of the present generation. The demand for skilled AI specialists is growing faster like never before. Requirements and open positions for experts in the sub-fields of AI like machine learning, deep learning, computer vision, statistics, and natural language processing are surging each day.

一个 rtificial智力是在当今发展最快的领域。 据《 财富 》杂志统计,人工智能专家的招聘在过去四年中增长了74%。 人工智能被视为当代的“最热门”工作。 对熟练的AI专家的需求以前所未有的速度增长。 人工智能子领域(如机器学习,深度学习,计算机视觉,统计和自然语言处理)的专家的要求和职位空缺每天都在增加。

What better time to make a startup of your own? You could be a self-made entrepreneur with some artificial intelligence skills and knowledge. Thanks to all the hype, there simply just can’t be a better time to build your very own startup in AI. However, this requires some fact-checking and digging up.

有什么更好的时间来自己创业? 您可以成为具有一些人工智能技能和知识的白手起家的企业家。 由于所有的炒作,在AI中建立自己的初创公司简直是一个更好的时机。 但是,这需要进行一些事实检查和挖掘。

Which fields actually live up to all the accelerated hype by AI? To answer this question we will have to explore the various fields of artificial intelligence and the various ideas that come along with each topic. I will be discussing the various possible startups with AI. Then I will recommend a few suggestions to get started with that particular startup and proceed to explain it in a bit more detail. Let’s start analyzing the 6 best startup choices for AI in no particular order.

哪些领域真正符合AI的所有加速宣传? 为了回答这个问题,我们将必须探索人工智能的各个领域以及每个主题附带的各种思想。 我将讨论使用AI的各种可能的创业公司。 然后,我将建议一些建议,以着手于该特定的启动程序,并继续对其进行详细说明。 让我们开始按特定顺序分析AI的6种最佳启动选择。

1.机器学习创业公司: (1. Machine Learning Startups:)

Ideas and approaches — Optimization algorithms to solve complex problems, high-quality recommendation systems, advanced spam filtering, fraud detection.

想法和方法-解决复杂问题的优化算法,高质量推荐系统,高级垃圾邮件过滤,欺诈检测。

Source: Photo by Daniel Wiadro on unsplash 来源: Daniel Wiadro摄影,未飞溅

Machine learning startups offer a wide range of opportunities. There are so many directions you can choose. This is also a great option for people with proficient machine learning knowledge but lack the deep learning perspective or are not that interested in deep learning. Some people just like more complex math and machine learning designing. Well lucky for them! This is the perfect startup option. The designing, strategic planning, and deployment of various machine learning models can be performed on the idea of their choice to earn cash and make profits on their business venture.

机器学习创业公司提供了广泛的机会。 您可以选择很多方向。 对于拥有精通机器学习知识但缺乏深度学习观点或对深度学习不感兴趣的人来说,这也是一个不错的选择。 有些人喜欢更复杂的数学和机器学习设计。 他们很幸运! 这是完美的启动选项。 各种机器学习模型的设计,战略规划和部署都可以根据他们选择的构想来进行,以从他们的企业中赚钱和获利。

2.自然语言处理: (2. Natural Language Processing:)

Ideas and approaches — Unique innovative chatbots for business and industry, sequence to sequence models with attention for machine translation for regional languages, text to speech, and speech to text tasks.

想法和方法- 独特的创新型聊天机器人,适用于工商业,采用点对点模型,并专注于区域语言,文本到语音以及语音到文本任务的机器翻译。

Source: Photo by Kourosh Qaffari on unsplash 资料来源: Kourosh Qaffari摄影,未飞溅

Do you love text, lots of reading, pre-processing, and building models with text-based data? Bingo! This is the best choice to plan and build your startup on. The possibilities you have with natural language processing as a startup choice are diverse. These range from classification of text, segregation of useful data, construction of high-level chatbots, and much more. If you live in a place where the regional language is not yet verified as an official language in the google translator, that could be a great business option as well. Also, most google translations do not work well on some languages. You can improvise your own datasets and ideas resulting in a successful startup. If you want to build chatbots, the best approach is to look for what are your target audience, companies, or businesses. Making specific chatbots is ideal as you can greatly improve the performance for that particular task.

您喜欢文本,大量阅读,预处理以及使用基于文本的数据构建模型吗? 答对了! 这是计划和构建启动的最佳选择。 使用自然语言处理作为启动选择的可能性是多种多样的。 这些范围包括文本分类,有用数据的隔离,高级聊天机器人的构造等等。 如果您居住的地方还没有在google翻译器中确认为官方语言,那么这也是一个不错的业务选择。 另外,大多数Google翻译在某些语言上无法正常运行。 您可以临时创建自己的数据集和想法,从而成功启动。 如果要构建聊天机器人 ,最好的方法是寻找目标受众,公司或企业。 制作特定的聊天机器人非常理想,因为您可以大大提高特定任务的性能。

3.计算机视觉: (3. Computer Vision:)

Ideas and approaches — Face recognition, object detection, human activity recognition, human emotion and gesture detection, image segmentation, optical character recognition (OCR).

想法和方法— 面部识别 ,对象检测,人类活动识别, 人类情感和手势检测 ,图像分割,光学字符识别(OCR)。

Source: Photo by Carl Heyerdahl on unsplash 资料来源: Carl Heyerdahl摄,未飞溅

I am personally very pumped for computer vision projects and computer vision related startups. This field offers unique and diverse options to build so many innovative projects. Computer vision projects are meant to interpret and visualize the images and objects in the surrounding. A fruitful computer vision business venture can generate a high income. Computer vision can be used across many platforms. Face recognition tasks can be performed on smartphones, IoT devices, etc. Object detection tasks can be performed for image retrieval, surveillance, automated vehicles, and machine inspection. Here is an example of human emotion and gesture recognition by me. Set up your startup by picking your target audience and company for which you want to perform the specific computer vision task.

我个人非常喜欢计算机视觉项目和与计算机视觉相关的初创公司。 该领域提供了独特而多样的选择来构建如此众多的创新项目。 计算机视觉项目旨在解释和可视化周围的图像和对象。 富有成效的计算机视觉企业可以产生高收入。 计算机视觉可以在许多平台上使用。 面部识别任务可以在智能手机,IoT设备等上执行。对象检测任务可以在图像检索,监视,自动车辆和机器检查中执行。 是我对人类情感和手势识别的示例。 通过选择要为其执行特定计算机视觉任务的目标受众和公司来设置启动。

4.基于医疗AI的初创企业: (4. Medical AI-based Startups:)

Ideas and approaches — Advanced x-ray scanning, advanced screening and diagnosis of malicious diseases, retinal scans, identification of various symptoms for specific conditions.

想法和方法-先进的X射线扫描,先进的恶意疾病筛查和诊断,视网膜扫描,针对特定情况的各种症状的识别。

Source: Photo by Owen Beard on unsplash 资料来源: Owen Beard摄影,未飞溅

Artificial intelligence is being extensively used in the medical research and science fields. Medical startups with artificial intelligence technologies are gaining extremely high popularity due to the precision and accuracy of predictions. The development of high-quality models especially in the field of medicine and medical science is essential. This is because an individual’s life could be on the line with each prediction. This is one of the main reasons the demand for these startups are rising each day. This is the best startup idea for you if you are interested in helping people and innovating the equipment in the medical industry.

人工智能正在医学研究和科学领域中广泛使用。 凭借人工智能技术的医疗初创企业由于预测的准确性和准确性而获得了极大的欢迎。 开发高质量模型尤其是在医学和医学领域至关重要。 这是因为个人的生活可能与每项预测都息息相关。 这是这些创业公司的需求每天都在增加的主要原因之一。 如果您有兴趣在医疗行业帮助人们和创新设备,这是最佳的启动想法。

5.带有AI的物联网(基于IoT的AI初创公司): (5. Internet of Things with AI (IoT-based AI Startups):)

Ideas and approaches — AI-based robotics, security using embedded systems, self-driving AI cars, face recognition and emotion analysis on embedded devices, home automation setups.

想法和方法-基于AI的机器人技术,使用嵌入式系统的安全性,自动驾驶AI汽车,嵌入式设备上的面部识别和情感分析,家庭自动化设置。

Source: Photo by Marvin Meyer on unsplash 资料来源: Marvin Meyer摄影,未飞溅

This is perhaps the coolest startup idea with AI. Both IoT and AI are buzz-words in today’s world. There are tons of options one can choose to integrate AI with IoT. Incorporation of AI with embedded systems to perform various chores like security, surveillance, face recognition, and finger print recognition among many others are fantastic startup options. Robotics and AI are one of the most popular fields together. To read more about the AI-based robots dream check out this article here. Overall, this is the best option for anyone interested in real-life and real-time computational processing of IoT projects together with artificial intelligence.

这可能是AI最酷的启动想法。 物联网和人工智能都是当今世界的流行语。 人们可以选择无数种选择来将AI与IoT集成。 将AI与嵌入式系统集成在一起以执行各种繁琐的工作,例如安全性,监视,面部识别和指纹识别等,是绝佳的启动选项。 机器人技术和AI一起是最受欢迎的领域之一。 要了解有关基于AI的机器人梦想的更多信息,请在此处查看本文。 总体而言,对于那些对物联网项目的实时和实时计算处理以及人工智能感兴趣的人,这是最佳选择。

6.虚拟助手: (6. Virtual Assistants:)

Ideas and approaches — Speech recognition to interact with the user, conversational chatbot AI, natural language processing, computer vision knowledge.

想法和方法—与用户交互的语音识别,对话式聊天机器人AI,自然语言处理,计算机视觉知识。

Source: Photo by Bence Boros on unsplash 资料来源: Bence Boros摄影,未飞溅

Virtual assistants require a special mention of their own. This is because the more precise and robust virtual assistants are quite complex to build. However, the revenue generated by these virtual assistant startups can be very high. It is important to have some knowledge across IoT, natural language processing, computer vision, and speech translation. The existing popular virtual assistants are Amazon Alexa, Apple’s Siri, Google Assistant, and Microsoft’s Cortana. The market needs more virtual assistants. People love the options these virtual assistants provide. However, virtual assistants like Alexa can be slightly expensive for some people. This opens up an amazing startup idea to sell. Cheaper better performing virtual assistants with optimized algorithms can be a great selling point. I will be a doing series in the near future on how can you make your very own virtual assistant.

虚拟助手需要特别提及自己。 这是因为构建更精确,更强大的虚拟助手非常复杂。 但是,这些虚拟助手初创公司产生的收入可能非常高。 拥有有关物联网,自然语言处理,计算机视觉和语音翻译的知识非常重要。 现有流行的虚拟助手是Amazon Alexa,Apple的Siri,Google Assistant和Microsoft的Cortana。 市场需要更多的虚拟助手。 人们喜欢这些虚拟助手提供的选项。 但是,对于某些人来说,像Alexa这样的虚拟助手可能会稍微贵一些。 这开辟了一个惊人的创业想法来出售。 使用优化算法的廉价,性能更好的虚拟助手可能是一个不错的卖点。 我将在不久的将来做一个系列文章,介绍如何制作自己的虚拟助手

Source: Photo by V2osk on unsplash 资料来源: V2osk摄影,未飞溅

最后的想法: (Final Thoughts:)

A Character cannot be developed in ease and quiet. Only through experience of trial and suffering can the soul be strengthened, ambition inspired, and success achieved.

角色不能轻松安静地发展。 只有通过经历磨难和磨难的经验,才能使灵魂得到加强,激发雄心并取得成功。

Helen Keller

海伦·凯勒

These are overall the 6 best startup ideas in my experience. Feel free to let me know if you felt I missed something or you feel any other startup idea is on the rise. At the end of the day, self-satisfaction and feeling happy about what you perceive is the most important thing in life. If you do choose any of the following startup ideas, I wish you all the best of luck! Enjoy your startup idea without the fear of success or failure.

这些是我经验中最好的6个最佳启动创意。 随时让我知道您是否觉得我错过了什么,或者您觉得其他任何新兴想法正在兴起。 归根结底,自我满足和对自己的感知感到高兴是生活中最重要的事情。 如果您选择以下任何一种启动方法,希望您一切顺利! 享受您的创业想法,而不必担心成功或失败。

The real test is not whether you avoid this failure because you won’t. It’s whether you let it harden or shame you into inaction, or whether you learn from it; whether you choose to persevere. — Barack Obama

真正的考验不是您是否会避免失败,因为您不会。 这是您让它变得僵硬还是让自己无所作为而感到羞耻,还是您是否从中学到东西? 是否选择坚持不懈。 - 巴拉克奥巴马

Setting up your startup idea and a concrete plan is crucial for the prosperity of your startup and the journey to becoming a successful entrepreneur. I hope all of had fun reading this article and were able to learn a few things from this. Good luck to all the people setting out to be AI entrepreneurs of the modern era! Have a wonderful day and see you all in another article.

设定创业理念和具体计划对于创业的繁荣和成为成功企业家的旅程至关重要。 我希望所有人都阅读本文并从中学习到一些东西。 祝所有准备成为现代AI企业家的人们好运! 祝您有美好的一天,在另一篇文章中见。

翻译自: https://towardsdatascience.com/6-best-startup-choices-with-artificial-intelligence-360260ccf89e

ai人工智能

本文内容由网友自发贡献,版权归原作者所有,本站不承担相应法律责任。如您发现有涉嫌抄袭侵权的内容,请联系:hwhale#tublm.com(使用前将#替换为@)

ai人工智能_人工智能的6种最佳启动选择 的相关文章

  • Hazelcast:连接到远程集群

    我们有一组 Hazelcast 节点 全部运行在一个远程系统 具有许多节点的单个物理系统 上 我们希望从外部客户端连接到该集群 一个 Java 应用程序 它使用如下代码连接到 Hazelcast ClientConfig clientCon
  • .pyx 文件出现未知文件类型错误

    我正在尝试构建一个包含 pyx 文件的 Python 包 pyregion 但在构建过程中出现错误 检查以下输出 python setup py build running build running build py creating b
  • Java8 项目上的 SonarQube 给出 jacoco-Exception

    我刚刚下载了最新版本 SonarQube 4 3 然后尝试使用以下命令构建 java 8 项目 mvn clean install mvn sonar sonar 这给了我下面的例外 谷歌搜索 我的印象是这是一个早期的问题 应该已经解决 h
  • 使用 python 绘制正值小提琴图

    我发现小提琴图信息丰富且有用 我使用 python 库 seaborn 然而 当应用于正值时 它们几乎总是在低端显示负值 我发现这确实具有误导性 尤其是在处理现实数据集时 在seaborn的官方文档中https seaborn pydata
  • Tensorflow 与 Keras 的兼容性

    我正在使用 Python 3 6 和 Tensorflow 2 0 并且有一些 Keras 代码 import keras from keras models import Sequential from keras layers impo
  • 通过索引访问Python字典的元素

    考虑一个像这样的字典 mydict Apple American 16 Mexican 10 Chinese 5 Grapes Arabian 25 Indian 20 例如 我如何访问该字典的特定元素 例如 我想在对 Apple 的第一个
  • SDK尚未初始化,请务必先调用FacebookSdk.sdkInitialize()

    我在实现 Facebook SDK 时遇到此错误 并且我tried https stackoverflow com questions 15490399 error inflating class com facebook widget l
  • 使用 Keras np_utils.to_categorical 的问题

    我正在尝试将整数的 one hot 向量数组制作为 keras 将能够使用的 one hot 向量数组来拟合我的模型 这是代码的相关部分 Y train np hstack np asarray dataframe output vecto
  • ANTLR 获取并拆分词法分析器内容

    首先 对我的英语感到抱歉 我还在学习 我为我的框架编写 Python 模块 用于解析 CSS 文件 我尝试了 regex ply python 词法分析器和解析器 但我发现自己在 ANTLR 中 第一次尝试 我需要解析 CSS 文件中的注释
  • 快速像素绘图库

    我的应用程序以每像素的方式生成 动画 因此我需要有效地绘制它们 我尝试过不同的策略 库 但结果并不令人满意 尤其是在更高分辨率的情况下 这是我尝试过的 SDL 好的 但是慢 OpenGL 像素操作效率低下 xlib 更好 但仍然太慢 svg
  • 在 keras 中保存和加载权重

    我试图从我训练过的模型中保存和加载权重 我用来保存模型的代码是 TensorBoard log dir output model fit generator image a b gen batch size steps per epoch
  • Mac OSX 10.6 上的 Python mysqldb 不工作

    我正在使用 Python 2 7 并尝试让 Django 项目在 MySQL 后端运行 我已经下载了 mysqldb 并按照此处的指南进行操作 http cd34 com blog programming python mysql pyth
  • Anaconda 无法导入 ssl 但 Python 可以

    Anaconda 3 Jupyter笔记本无法导入ssl 但使用Atom终端导入ssl没有问题 我尝试在 Jupyter 笔记本中导入 ssl 但出现以下错误 C ProgramData Anaconda3 lib ssl py in
  • Javac 版本 1.7 无法为目标 1.7 构建

    我试图在 Linux Mint 系统上使用 Sun Java JDK 1 7 0 17 编译 Java 代码 但遇到了这个问题 javac version target 1 7 javac 1 7 0 17 javac invalid ta
  • 设置 Firefox 配置文件以使用 Selenium 和 Java 自动下载文件

    我想使用 Selenium WebDriver 和 Java 验证文件下载 要下载的文件为 PDF 格式 当 WebDriver 单击 AUT 中的 下载 链接时 Firefox 将打开以下下载确认窗口 我希望 Firefox 自动下载文件
  • 乔达时间中两个日期之间的天数

    如何找到两次之间的天数差异乔达时间 http www joda org joda time DateTime http www joda org joda time apidocs org joda time DateTime html实例
  • while循环只执行一次

    我很难弄清楚为什么 while 循环实际上不会循环 它运行一次并停止 import java util public class mileskm public static void main String args Scanner inp
  • 多个对象以某种方式相互干扰[原始版本]

    我有一个神经网络 NN 当应用于单个数据集时 它可以完美地工作 但是 如果我想在一组数据上运行神经网络 然后创建一个新的神经网络实例以在不同的数据集 甚至再次同一组数据 上运行 那么新实例将产生完全错误的预测 例如 对 XOR 模式进行训练
  • 如何让JComboBox中的内容居中显示?

    目前我有这个JComboBox 我怎样才能将其中的内容居中 String strs new String 15158133110 15158133124 15158133458 JComboBox com new JComboBox str
  • Swing:创建可拖动组件...?

    我在网上搜索了可拖动 Swing 组件的示例 但我发现示例不完整或不起作用 我需要的是一个摇摆组件那可以是dragged通过鼠标 在另一个组件内 被拖拽的时候 应该已经 改变它的位置 而不仅仅是 跳 到目的地 我很欣赏无需非标准 API 即

随机推荐

  • oracle无法减小列长度,中国港湾扩展表单分配时调整字段后保存时报错,无法正常维护表单定义...

    版本号 7 0 4 业务场景 如下 中国港湾扩展表单分配时调整字段后保存时报错 无法正常维护表单定义 异常提示 异常信息 添加业务实体的时候出错 导致错误的应用程序或对象的名称 Genersoft Platform RuntimeADP C
  • NodeJS使用JWT

    JWT jsonwebtoken 目前最流行的跨域身份验证解决方案 在nodejs中使用 安装 npm install jsonwebtoken save 使用 1 在路由中引入 var jwt require jsonwebtoken 2
  • UML 类图

    参考地址 UML类图 简书 一 类图的属性 1 写法格式 属性 方法名 类型 2 权限符号 表示private 表示protected 表示default 也就是包权限 下划线表示static 斜体表示抽象 二 类的关系 1 泛化 Gene
  • 华为OD机试真题- 喊7的次数重排-2023年OD统一考试(B卷)

    题目描述 喊7是一个传统的聚会游戏 N个人围成一圈 按顺时针从1到N编号 编号为1的人从1开始喊数 下一个人喊的数字为上一个人的数字加1 但是当将要喊出来的数字是7的倍数或者数字本身含有7的话 不能把这个数字直接喊出来 而是要喊 过 假定玩
  • Nowcoder专项练习:操作系统(二)

    1 动态重定位分区分配算法 虽然动态分区法比固定分区法的内存利用率高 但它还是有零头 碎片 的问题 动态重定位分区分配算法与动态分区分配算法基本上相同 差别仅在于 在这种分配算法中 添加了紧凑功能 通常 在找不到足够大的空闲分区来满足用户需
  • 高频测试面试题

    测试面试题 1 软件测试的目的与原则是什么 答 目的 A 通过测试工作可以发现并修复软件当中存在的缺陷 B 可以降低同产品开发遇到的风险 C 记录软件运行过程中的一些数据 从而为决策者提供技术支持 原则 A 2 8定律 核心功能占20 非核
  • Unity3d 相机平滑的跟随角色

    C 脚本 代码如下 示例 using System Collections using System Collections Generic using UnityEngine public class CameraFollow MonoB
  • Oracle中Clob字段类型修改为Varchar2类型实例

    新建一个字段 alter table KQSERVER add temp VARCHAR2 2000 将原来字段数据赋值到新字段 update KQSERVER set temp SERVER IMG 删除原来字段 alter table
  • 【华为OD机试】按身高和体重排队【2023 B卷

    华为OD机试 真题 点这里 华为OD机试 真题考点分类 点这里 题目描述 某学校举行运动会 学生们按编号 1 2 3 n 进行标识 现需要按照身高由低到高排列 对身高相同的人 按体重由轻到重排列 对于身高体重都相同的人 维持原有的编号顺序关
  • Unity shader 笔记之迷雾效果

    先看效果图 Shader Unlit FogOfWar Properties MainTex Texture 2D white CGINCLUDE include UnityCG cginc struct appdata float4 ve
  • 各国搜索引擎大全

    找客户做seo都要用到搜索引擎 全球除了谷歌 每个国家都有自己的百度 用翻译软件翻译成当地的语言搜索可能会有惊喜 中国百度 http www baidu com等 菲律宾 Philippine http www yehey com 韩国 K
  • 【08】Nginx之负载均衡

    负载均衡的作用 1 解决服务器的高并发压力 提高应用程序的处理性能 2 提供故障转移 实现高可用 3 通过添加或减少服务器数量 增强网站的可扩展性 4 在负载均衡器上进行过滤 可以提高系统的安全性 实现方式 方式一 DNS轮询方式 域名系统
  • 二、Dubbo+SpringBoot+zookeerper整合(注解方式)

    Git地址 https github com boorZ demo dubbo git 注意 Git地址上的项目与本文章有差同 请参照文章与Git Git地址 https github com boorZ demo dubbo xml 注意
  • QTableView的表格项中加入图标的方法

    当在使用表格视图的时候 需要在表格每一行前面加入图标 应该怎么做呢 Qt中通过使用MVC的处理方式 很容易做到这一点 具体实现如下 先贴出图 让大家一睹为快 下面我就来介绍一下 上图的灯泡是怎么实现的 通过重载QAbstractTableM
  • 如何安装并运行SiB-CROP模型

    目录 1 环境配置 1 1 Supported Platforms 1 2 Prerequisites 2 下载SIB CROP代码 3 SIB CROP模型编译 4 下载驱动数据 5 SIB CROP模型运行 主要参考 1 官方文档 ht
  • PSYoungGen ParNewGeneration DefNewGeneration等名词解释

    HotSpot VM的GC组老人之一Jon Masamitsu很久之前就写过blog讲解这个 https blogs oracle com jonthecollector entry our collectors 简单来说 有这么多东西反映
  • Python+opencv:图像修复

    简介 OpenCV 是一个开源的计算机视觉库 它包含了许多图像处理和计算机视觉算法 使用 OpenCV 进行图像修复主要依赖于传统的图像处理技术 OpenCV 图像修复方法及其原理 1 去噪 图像去噪是消除图像中的噪声 提高图像质量的过程
  • Springboot2.0 上传图片 jar包导出启动(第二章)

    目录 一 目录文件结构讲解 二 文件上传实战 三 jar包方式运行web项目的文件上传和访问处理 核心知识 最后 一 目录文件结构讲解 简介 讲解SpringBoot目录文件结构和官方推荐的目录规范 1 目录讲解 src main java
  • 连接已失效_西门子S7-1500PLC如何通过IP地址进行HMI连接

    通过系统 IP 地址进行 HMI 连接 要求 S7 1500R H 冗余系统 如 CPU 1513R 1PN 系统 IP 地址已启用 带有 PROFINETI 接口的 HMI 设备 操作步骤 要与 S7 1500R H 冗余系统建立 HMI
  • ai人工智能_人工智能的6种最佳启动选择

    ai人工智能 意见 Opinion Artificial Intelligence is the fastest growing field in the present day According to fortune the stati